Observational Studies of Inhaled Corticosteroid Effectiveness in COPD: Lessons Learned

Area of Science:

• Pulmonary Medicine
• Clinical Epidemiology
• Pharmacovigilance

Background:

• Observational studies have suggested inhaled corticosteroids (ICS) reduce mortality in COPD.
• Randomized controlled trials, like the TORCH trial, have not consistently supported these findings.
• Discrepancies between observational data and trial results highlight potential issues in study design.

Purpose of the Study:

• To systematically review observational studies on ICS effectiveness in COPD.
• To identify and analyze biases affecting the results of these observational studies.
• To determine the true effect of ICS on COPD outcomes like death and hospitalization.

Main Methods:

• A comprehensive literature search was conducted for observational studies (cohort, case-control) on ICS in COPD.
• Studies were assessed for specific biases, including immortal time bias, immeasurable time bias, and the "asthma factor" bias.
• Meta-analysis was performed on studies with and without identified biases.

Main Results:

• Twenty-one observational studies were identified, with most affected by significant biases.
• Biased studies suggested substantial reductions in COPD mortality and hospitalization with ICS use.
• Five studies free from major biases found no association between ICS and improved COPD outcomes (pooled rate ratio, 1.02).

Conclusions:

• Observational studies are valuable for real-world data but require rigorous design to avoid bias.
• Biases in observational studies may overestimate the benefits of ICS in COPD.
• Careful review of observational data is crucial before initiating large randomized trials for COPD treatments.