A clinical trial on back pain was carried out with two
randomized treatment groups, intercostal/epidural (trt=1) and
morphine infusion (trt=2). Patients were asked to give an assessment
of their state of disturbance or discomfort on a Verbal Rating Scale
(possible values integers 1=least through 5=most) for back pain and 
on a Visual Analogue Scale (possible values integers 1=least through 
100=most) for anxiety and alertness. Measurements were recorded at four 
roughly equally spaced time points after treatment for each patient. 

Some analyses could use a dichotomized version of the outcome variable 
for back pain, for example by creating a new variable which is 1 if 
the VRS value is 3 or more and 0 otherwise. Covariates available 
include treatment, anxiety, and alertness. Note that anxiety and 
alertness are also measured at every time point.  The time points at
which the measurements were taken are given as minutes after treatment.

Data source: Hand and Crowder (1996). Practical Longitudinal Data
Analysis. London: Chapman and Hall. (Table A19)