Pharmacology Example

Pharmacology Example

Sample Size for Contrast in Anova

What sample size is required for an experiment to determine the nature of a dose response to a new drug? The study plan calls for a placebo group, and four doses of the new drug equally spaced on a log scale. Prior research with the outcome variable suggests that the within-group standard deviation is about 4.0, and that the placebo response will be about 16.0; the drug will be of interest if some dose produces a response of at least 20.0.

nQuery Advisor® can provide sample size and power calculations for the one-way analysis of variance, and for any specified contrasts, such as contrasts comparing all or any of the drug doses to the placebo, or contrasts to evaluate whether there is a linear or perhaps a quadratic dose response.

Select File … New, and in the Study Goal and Design Box, select Means, >Two Groups, and Test and select Single one-way contrast. After filling in the desired significance level and the number of dose groups being studied, select Compute Effect Size from the Assistants menu or click on the button marked δ to specify the contrast coefficients and the expected means for every group, and to transfer these results back to the main sample size table.

You can specify the sample size for each group and solve for power for the test of the contrast, or specify the power required for the contrast and solve for the required sample size per group. Here we specify a linear contrast in log dose, and ask for the sample size per group required for 90% power assuming that there is a linear dose response and only the highest dose produces the required mean of 20.

For this example, a sample of 35 subjects per group would be required for a total sample size of 175. You can evaluate sample sizes required for other hypothesized sets of means and/or other contrasts in the additional table columns. If the study design calls for unequal sample sizes in each group, you can request tables allowing specification of different sample sizes in each group and solve for power for specified group sample sizes or solve for total sample size given power and the ratios of sample sizes in each group.
nQuery Advisor® provides tables, plots, and standardized sample size justification statements which can be printed directly or copied to clipboard and pasted into your manuscript or proposal.