headache: Randomized crossover study comparing two analgesic drugs (A...
In ALA: Data sets and examples for the book "Applied Longitudinal Analysis"

Description Usage Format Details Note Source References Examples

The study was designed to compare two active drugs and placebo for relief of tension headache.

1	data(headache)

A data frame with 846 observations on the following 6 variables.

id: a factor with 423 levels
center: a factor with 14 levels
treatment.seq: a factor with levels AB, BA, AP, PA, BP, PB
period: a factor with levels 0 and 1
treatment: a factor with levels A, B, and P
relief: a numeric vector

The two analgesic drugs were identical in their active ingredients except that one contained caffeine. The primary comparison of interest was between the two active drugs; the placebo was included for regulatory purposes. Note that there were three treatments, but only two periods, i.e., each subject received only two of the three treatments in a random order. With three treatments and two periods, there are six possible treatment sequences, AB, BA, AP, PA, BP, PB, where A, B and P denote the two active drugs and placebo. In this study the AB and BA sequences were assigned three times as many subjects as the remaining four because of the interest in the A versus B comparison.

Original variable names have been adapted to R conventions. The third column in the source was excluded as it was a duplicate of the sixth. The center ID was prefixed with the letter “C” in the center variable, compared to the source.

Two headaches treated within each period and response is the average pain relief for both headaches.

http://biosun1.harvard.edu/~fitzmaur/ala

Laird NM, Skinner J, Kenward M (1992) An analysis of two-period crossover designs with carry-over effects. Statistics in Medicine 11:1967-1979