Table 2 |
||||
|
Illustrating how the "Intention to treat" approach ascribing indeterminate treatment outcomes as failures overestimates the true failure rate. Low failure rate |
||||
| Follow-up |
A (%) |
R = F(%) |
FITT (%) |
Overestimation of failure rate (%) |
|
|
||||
| 6 weeks |
6 |
3 |
3.3 |
10% |
| 8 weeks |
20 |
5 |
6 |
20% |
| 10 weeks |
45 |
5 |
7.25 |
45% |
| 12 weeks |
68 |
5 |
8.4 |
68% |
| 20 weeks |
75 |
5 |
8.75 |
75% |
|
|
||||
|
If the failure rate is low (as it hopefully should be) the errors using the ITT approach are very large. In this example everything is the same as in the above example but now the true failure rate (F) is 5%, A – cumulative probability of developing a patent new infection R – cumulative probability of developing a patent recrudescence |
||||
|
Stepniewska and White Malaria Journal 2006 5:127 doi:10.1186/1475-2875-5-127 |
||||
