Systematic reviews do not adequately report or address missing outcome data in their analyses: a methodological...

Area of Science:

• Medical Research Methodology
• Evidence Synthesis
• Biostatistics

Background:

• Missing outcome data is a common challenge in clinical trials.
• Systematic reviews synthesize evidence from multiple trials, making consistent reporting of missing data crucial.
• Inconsistent reporting of missing data can bias systematic review results.

Purpose of the Study:

• To assess how systematic review authors report and handle missing outcome data from trial participants.
• To evaluate the transparency and methodological rigor in addressing missing data within systematic reviews.

Main Methods:

• A methodological survey was conducted on a random sample of 100 systematic reviews (50 Cochrane, 50 non-Cochrane) that reported group-level meta-analyses.
• Reviews were analyzed for reporting of missing outcome data categories in methods and results sections.
• The study examined methods used to handle missing data and conduct sensitivity analyses.

Main Results:

• Only 25% of reviews planned to address missing data categories, while 42% reported some data on missingness.
• "Unexplained loss to follow-up" was the most frequently reported category.
• Few reviews (19%) used methods beyond complete case analysis for primary analyses, and only 9% reported sensitivity analyses for missing data bias.

Conclusions:

• Most systematic reviews lack sufficient reporting on missing outcome data categories.
• There is a general inadequacy in addressing missing data within the primary analyses of systematic reviews.
• Improved reporting and handling of missing data are necessary to enhance the reliability of evidence synthesis.