Introductory Example

We consider a very simple cross-over trial using artificial data. In this trial each patient receives each of treatments A and B for a fixed period of time. At the end of each treatment period, a measurement is taken to assess the response to that treatment. In the analysis of such a trial we commonly refer to treatments being crossed with patients, meaning that the categories of 'treatments' occur in combination with the categories of 'patients'. For the purpose of this illustration we will suppose that the response to each treatment is unaffected by whether it is received in the first or second period. The table shows the results from the six patients in this trial.

Treatment

Patient

A

B

Difference A - B

Patient mean

1

20

12

8

16.0

2

26

24

2

25.0

3

16

17

-1

16.5

4

29

21

8

25.0

5

22

21

1

21.5

6

24

17

7

20.5

Mean

22.83

18.67

4.17

20.75

Was this article helpful?

0 0

Post a comment