Hierarchy of Strength of Evidence or Interpretability

Let P_i denote patient i and the treatments be denoted by A and B. Thus P₂^B represents patient 2 on treatment B. P₁ represents the average outcome over a sample of patients from which patient 1 was selected.

Design	Patients Compared

6-period crossover	P₁^A vs P₁^B (directly measure HTE)
2-period crossover	P₁^A vs P₁^B
RCT in idential twins	P₁^A vs P₁^B
∥ group RCT	P₁^A vs P₂^B, P₁=P₂ on avg
Observational, good artificial control	P₁^A vs P₂^B, P₁=P₂ hopefully on avg
Observational, poor artificial control	P₁^A vs P₂^B, P₁≠ P₂ on avg
Real-world physician practice	P₁^A vs P₂^B