systematic-review-epi

To perform meta-analysis in Python, this Skill pools study-level effect sizes and standard errors into fixed- and random-effects models, generating combined estimates and matplotlib forest plots for visualizing pooled outcomes.

Heterogeneity quantification calculates Cochran's Q and I² statistics to measure between-study variability in systematic reviews. This Skill computes these metrics natively using scipy and statsmodels to assess whether study differences exceed chance.

Yes, you can detect publication bias by applying Egger's test, Begg-style rank correlation tests, and generating funnel plots. This Skill runs these statistical diagnostics to identify asymmetry indicative of unpublished negative findings.

The best way to explore effect modifiers is through meta-regression, which tests whether study-level covariates like setting or year explain heterogeneity. This Skill fits regression models to assess covariate influence on pooled outcomes.

You need Python with pandas, numpy, and scipy to compute the required statistics for GRADE evidence synthesis. This Skill uses these libraries to produce a structured quality rating of the overall evidence.

Use random-effects models when heterogeneity quantification shows significant between-study variability, and fixed-effects when studies are functionally identical. This Skill computes both to ensure heterogeneity-aware pooled estimates.