Emulating randomized trials by observational database studies: the RCT-DUPLICATE initiative in COPD and asthma

Area of Science:

• Pharmacoepidemiology
• Real-world evidence generation
• Comparative effectiveness research

Background:

• Observational studies are increasingly vital for regulatory decision-making, providing real-world evidence.
• The RCT-DUPLICATE initiative aimed to assess the consistency between observational studies and randomized trials.

Purpose of the Study:

• To emulate published randomized trials using observational data in asthma and chronic obstructive pulmonary disease (COPD).
• To compare treatment effects derived from emulated observational studies versus their original randomized trials.

Main Methods:

• New-user cohorts were created from US healthcare claims databases for asthma and COPD patients.
• Propensity score matching was employed to balance baseline characteristics between treatment groups.
• Proportional hazards models were utilized to estimate hazard ratios for key outcomes.

Main Results:

• Observational studies included larger patient populations than the original trials.
• Results for asthma (serious asthma-related events) and COPD (exacerbations) showed discordance between observational analyses and randomized trials.
• Hazard ratios for asthma: 1.29 (observational) vs. 1.07 (trial); for COPD: 1.08 (observational) vs. 0.84 (trial).

Conclusions:

• Significant discordance was observed between emulated observational studies and their corresponding randomized trials.
• Differences may be attributed to trial designs, such as forced treatment discontinuation before randomization, which cannot be replicated in observational data.
• This highlights potential challenges in directly comparing real-world evidence from observational studies with findings from traditional randomized trials.