A mixed effects model for the analysis of repeated measures cross-over studies

A mixed effects model is developed for cross-over trials in which the respo nse is measured repeatedly within each time period. Relative to previous wo rk on repeated measures cross-overs, the methodology synthesizes two import ant features. First, our procedure eliminates preliminary testing for carry -over, defined loosely as the component of a response that is due to treatm ent in the preceding period. This is achieved by generalizing the methodolo gy to cross-over designs in which preliminary testing for carry-over is unn ecessary. We focus largely on 'simple' carry-over, that is, carry-over that lasts for exactly one period and is independent of the treatment administe red in the period in which the carry-over occurs. However, we also illustra te a modification of the procedure for a repeated measures cross-over desig n which uses a more complicated model of carryover. Second, the model allow s both the between- and within-subject variance to differ among treatments. Conditions are described wherein closed-form (CF) solutions to the varianc e components as well as closed-form hypothesis tests of the treatment diffe rences exist. Flexibility in the model is illustrated with an example in wh ich inference based on the CF likelihood-based estimates of the variance, a nd estimates formed using an iterative routine (PROC MIXED) are compared. C opyright (C) 1999 John Wiley & Sons, Ltd.