1.   None the less, the sharp differences in survival curves between the three comorbidity groups indicate that this simple categorisation was valid.

2.   Survival curves were analysed using the actuarial method of Mantel-Haenzel.

3.   Survival curves of the three histological groups were analysed according to the Kaplan and Meier method and the differences between these curves were calculated with the Mantle test.

4.   Lifetable analysis of the truncated survival curve also indicates that patients taking cisapride were less likely to develop a recurrent ulcer over the study period than those receiving placebo.

5.   Survival curves were compared using a log rank test.

6.   The survival curve for the study shows that the highest mortality is during the first four years after diagnosis.

7.   Such survival curves take no account of the age structure of the patients.

8.   Thereafter the two survival curves are identical.

9.   Nonetheless, the sharp differences in survival curves between the three comorbidity groups indicate that this simple categorisation was valid.

10.   The survival curve between those who received the gene therapy and those who did not was basically no different.