The Impact of Study Size on COVID-19 Treatment Outcomes: A Meta-Epidemiological Study Comparing Large and Small...

Area of Science:

• Medical Research Methodology
• Epidemiology
• Biostatistics

Background:

• Small randomized controlled trials (RCTs) in COVID-19 meta-analyses are linked to favorable effects and instability.
• Assessing the impact of trial size on effect estimates and bias is crucial for reliable scientific conclusions.

Purpose of the Study:

• To evaluate how randomized controlled trial (RCT) size influences effect estimates, statistical stability, and risk of bias in COVID-19 meta-analyses.
• To compare findings from small versus large RCTs to inform meta-analysis practices.

Main Methods:

• Systematic review and meta-analysis of COVID-19 treatment RCTs following PRISMA guidelines.
• Trials classified by sample size (large vs. small); separate meta-analyses conducted.
• Comparative metrics included Ratio of Odds Ratios (ROR), Fragility Index (FI), Reverse Fragility Index (RFI), and risk of bias assessments.

Main Results:

• Small trials produced more extreme estimates (19/25 analyses) and wider confidence intervals (23/25 analyses).
• Pooled ROR was 0.85, decreasing to 0.81 when limited to small trials, indicating overestimation of treatment effects.
• Large trials showed greater stability (FI: 14.0 vs. 4.0) and agreement with overall estimates compared to small trials.

Conclusions:

• Small RCTs in COVID-19 meta-analyses tend to overestimate treatment effects and provide less precise, less stable results.
• Prioritizing large, high-quality RCTs in meta-analyses is recommended.
• Findings from small studies require cautious interpretation, especially in rapidly evolving research areas.