Table 1 Designs, hypotheses per trial, and possibility of ensuring equal Hawthorne effect across the arms.

Arm I vs. II Complete intervention on lot A. Using lot B as control.

Block

If the test scores related to lot A vignettes increase in accordance with Clinical Evidence chapters A and there is no change in test scores of B chapters, the intervention has a genuine effect.

No

Arm II vs. I Complete intervention on lot B. Using lot A as control.

Block

If the test scores related to lot B vignettes increase in accordance with Clinical Evidence chapters B and there is no change in test scores of A chapters, the intervention has a genuine effect.

No

Arm I vs. III Complete intervention for lot A (ECCE). Minimal intervention on A-B chapters (Clinical Evidence) (control).

Classical

If the test scores related to A vignettes in the intervention arm increase in accordance with the Clinical Evidence chapters A and there is no change in the control arm the intervention (ECCE) has a favourable effect, but without controlling for the Hawthorne effect.

Yes

Arm II vs. III Complete intervention for lot B (ECCE). Minimal intervention on A-B chapters (Clinical Evidence) (control).

Classical

If the test scores related to lot B vignettes in the intervention arm increase in accordance with the Clinical Evidence chapters B and there is no change in the control arm the intervention (ECCE) has a favourable effect, but without controlling for the Hawthorne effect.

Yes