Metasens: Statistical Methods for Sensitivity Analysis in Meta-Analysis

Metasens implements six statistical methods for sensitivity analysis in meta-analysis, supporting the work of Schwarzer et al. (2015). The methods include the Copas selection model, limit meta-analysis, and imputation methods for missing binary data. The package is authored by Guido Schwarzer.

Use Cases

• Evaluating sensitivity to publication bias based on the Copas selection model
• Assessing small-study effects using the LFK index test and Doi plot
• Imputing missing binary outcome data based on the Gamble & Hollis and Higgins et al. methods
• Estimating an upper bound for outcome reporting bias based on the Copas & Jackson method
• Applying limit meta-analysis to adjust for bias based on the Rücker et al. method

Strengths

• Implements six established statistical methods from peer-reviewed literature
• Directly supports a specific chapter (5) from Schwarzer et al. (2015)

Limitations

• Column-level documentation is absent; field semantics must be inferred after download
• Row count is unknown, which may limit suitability assessment

Provenance

Source

Guido Schwarzer

Collection Method

Software package implementing statistical methods.

Time Range

Methods range from 2001 to 2018.

Freshness

Last update date is unknown; freshness unverified

Geography

Global