dose-response

Dose-response meta-analysis using restricted cubic splines models non-linear relationships across studies by placing pre-specified knots, capturing complex curves that linear models miss. This approach fits summarized data and estimates trends using GLS.

Two-stage pooling for dose-response data first estimates study-specific curves independently, then combines them using multivariate random-effects models. This method accounts for within- and between-study variance, yielding a pooled non-linear dose-response curve.

To report a dose-response meta-analysis meeting PRISMA 2020 guidelines, document GLS trend estimation, spline knot placement, and pooling methods transparently. Include sensitivity analyses and follow structured reporting extensions for reproducibility.

Use GLS trend estimation for summarized dose-response data when modeling trends across multiple studies with correlated estimates. It applies generalized least squares to account for within-study covariance, providing robust slope estimates for pooling.

Yes, you can use dosresmeta for multivariate random-effects pooling in non-linear models. It facilitates two-stage pooling of study-specific restricted cubic spline curves, combining them while accounting for heterogeneity across systematic reviews.

Limitations of restricted cubic splines in dose-response meta-analysis include sensitivity to knot placement and potential overfitting with sparse data. If dose ranges vary drastically across studies, accurately estimating non-linear shapes becomes difficult.