Table 1.

Challenges in calibrating RWE against RCTs and measures of differences between the study types.

Emulation differences	Measures of emulation difference
Differences in study populations due to: • Unrecorded physician decision-making in RCT • Lack of complete patient histories in RWE • Inconsistent recording of laboratory values in RWE • Poor recording of patient symptoms in RWE	• Proportion of patients excluded due to each exclusion criterion in RCT vs RWE • Average available look-back in RWE • Proportion of patients with labs available in RWE
Differences in treatment strategies due to: • Placebo control in RCT • Tight control of medication adherence in RCT • Tight control of dosing schedule and treatment augmentation in RCT	• Proportion of patients discontinuing before the end of follow-up in RCT vs RWE • Proportion of patients reaching target dose during follow-up in RCT vs RWE • Proportion of patients on other therapies during follow-up in RCT vs RWE • Ordering of labs in RWE for treatment titration
Differences in outcome measurement due to: • Differing outcome definitions • Differential surveillance for outcomes • Differing lengths of follow-up	• Overall outcome risk in RCT vs RWE • Distributions of combined outcome events attributable to each cause in RCT vs RWE • Length of follow-up in RCT vs RWE • Evaluation of proportional hazards • Side-by-side Kaplan-Meier plots