Intention-to-treat analysis: difference between revisions
Diff·revision 2 → 3·05:54, 5 Sep 2024
Difference between revision 2 and revision 3 of Intention-to-treat analysis. 2 lines changed; the page grew by 200 bytes.
| Revision 2 — 21:41, 26 Aug 2024 FootnoteFyfe (talk) consistent en dashes in ranges 1,967 bytes ±0 | Revision 3 — 05:54, 5 Sep 2024 OpenLabelOtto (talk) state that the trial has reported but is not yet peer-reviewed 2,167 bytes +200 | ||
|---|---|---|---|
| 16 | The classic demonstration is that adherence itself predicts outcome even in a placebo arm: participants who take placebo reliably do better than those who do not, because adherence is a marker for other health-related behaviour. An analysis conditioning on adherence therefore finds an effect where none exists.{{r|hollis1999}} | 16 | The classic demonstration is that adherence itself predicts outcome even in a placebo arm: participants who take placebo reliably do better than those who do not, because adherence is a marker for other health-related behaviour. An analysis conditioning on adherence therefore finds an effect where none exists.{{r|hollis1999}} |
| 17 | 17 | ||
| + | 18 | The practical consequence is that a striking difference between an intention-to-treat and a per-protocol result is informative about differential dropout rather than about the treatment.{{r|ich_e9}} | |
| + | 19 | ||
| 18 | == References == | 20 | == References == |
| 19 | {{reflist}} | 21 | {{reflist}} |