Table 2.
Applying the Revised Bradford Hill guidelines
| Mother's kiss | Oral ulceration | ||
|---|---|---|---|
| Direct | 1. Size of effect not attributable to plausible confounding | Yes (dramatic effect; confounders highly unlikely) | Yes (dramatic effect; confounders highly unlikely) |
| 2. Appropriate temporal and/or spatial proximity | Yes (cure immediately follows the intervention and is spatially associated) | Yes (the effect is in immediate proximity to the intervention) | |
| 3. Dose‐responsiveness and reversibility | Not tested and not relevant (might have been tested by varying levels of expiratory force) | Not tested (dose‐responsiveness not tested; but subsequent healing suggested reversibility) | |
| Mechanistic | 4. Plausible mechanism of action | Yes | Yes (acidic compound) |
| 5. Coherence | Yes (nothing contradicts the causal hypothesis) | Yes (nothing contradicts the causal hypothesis) | |
| Parallel | 6. Replicability | Yes | Not tested |
| 7. Similarity | Not relevant | Yes (aspirin causes gastric erosions) | |
| Total | 5 ‘yes’ (1, 2, 4, 5, 6) 2 ‘not relevant’ or ‘not tested’ (3,7) | 5 ‘yes’ (1, 2, 4, 5, 7) 2 ‘not relevant’ or ‘not tested’ (3, 6) | |
| VERDICT | 5 out of 7 guidelines satisfied | 5 out of 7 guidelines satisfied |