The dosresmeta R package: Performing multivariate dose-response meta-analysis

The dosresmeta package is a free and open-source add-on for conducting (multivariate) dose-response meta-analysis with the statistical software environment R. You can install the package (or upgrade to the newest version) from CRAN directly within R by typing:

install.packages("dosresmeta")

Extensions to meta-regression models and alternative estimation procedures are included in an update version still under developement, which is avaiable on GitHub by typing:

install.packages("devtools")

devtools::install_github("alecri/dosresmeta")

 

Reproducible code

• GitHub repository for Crippa, A., Discacciati A., Bottai M., Spiegelman D., & Orsini N. “One-stage dose-response meta-analysis for aggregated data 2018. Statistical Methods in Medical Research

• GitHub repository for Crippa, A., & Orsini N. “Dose-response meta-analysis of differences in means” 2016. BMC Medical Research Methodology

• GitHub repository for Discacciati, A., Crippa, A., & Orsini N. “Goodness-of-fit tools for dose-response meta-analysis of binary outcomes” 2015. JRSM

 

Analysis Examples

The links below demonstrate how the models, methods, and techniques described in the respective articles/chapters can be applied via the dosresmeta package. The examples are run using dosresmeta version 2.0.0 avaiable on GitHub.

• Orsini, N. Weighted mixed-effects dose-response models for tables of correlated contrasts. Stata Journal. 2021, Vol.21 (2), p.320-347.

• Berlin JA, Longnecker MP, Greenland S. Meta-analysis of epidemiologic dose-response data. Epidemiology. 1993 May 1:218-28

• Orsini, N., Bellocco, R., & Greenland, S. Generalized least squares for trend estimation of summarized dose-response data. 2006 Stata Journal, 6(1), 40.

• Liu, Qin, et al. “A two-stage hierarchical regression model for meta-analysis of epidemiologic nonlinear dose–response data.” Computational Statistics & Data Analysis 53.12 (2009): 4157-4167.

 

Additional (useful) code

• Missing number of cases/n. Assuming independence
• Loop for selecting best fractional polynomial
• Leave-one-out dose-response meta-analysis
• Cumulative dose-response meta-analysis
• Sensitivity analysis to location of knots in a spline model

Older R script in txt format based on dosresmeta version 1.3.1 available on CRAN.

• Coffee consumption and all causes mortality
• Alcohol intake and colorectal cancer
• Alcohol intake and lung cancer
• Body mass index and renal cancer
• Alcohol intake and cardiovascular risk