Chapter 2 Introduction

This is a document that outlines a vignette for implementing privacy preserving survival models and meta-analyzing hazard ratios in the DataSHIELD platform.

We used the bookdown package (Xie 2022), R Markdown and knitr (Xie 2015) for this document. Our package dsSurvival (Banerjee and Bishop 2021a)(Banerjee and Bishop 2021b)(Banerjee et al. 2022) uses the metafor package for meta-analysis (Viechtbauer 2010).

2.1 Survival models

Survival models are used extensively in healthcare. Previously building survival models in DataSHIELD involved building piecewise exponential regression models. This is an approximation and involves having to define appropriate time buckets. A lack of familiarity with this approach also makes people suspicious.

The scope of our package implementation is restricted to being study-level meta-analysis (SLMA) rather than full likelihood.

References

Banerjee, Soumya, and Tom Bishop. 2021a. “neelsoumya/dsSurvivalClient: Survival functions (client side) for DataSHIELD.” https://doi.org/10.5281/zenodo.4806588.

Banerjee, Soumya, and Tom R. P. Bishop. 2021b. neelsoumya/dsSurvival: v1.0.0 Survival models in DataSHIELD (version v1.0.0). Zenodo. https://doi.org/10.5281/zenodo.4917552.

Banerjee, Soumya, Ghislain Sofack, Thodoris Papakonstantinou, Demetris Avraam, Paul Burton, Daniela Zöller, and Tom RP Bishop. 2022. “dsSurvival: Privacy preserving survival models for federated individual patient meta-analysis in DataSHIELD.” bioRxiv, January, 2022.01.04.471418. https://doi.org/10.1101/2022.01.04.471418.

Viechtbauer, Wolfgang. 2010. “Conducting Meta-Analyses in R with the metafor Package.” J. Stat. Softw. 36 (3): 1–48. https://doi.org/10.18637/jss.v036.i03.

Xie, Yihui. 2015. Dynamic Documents with R and Knitr. 2nd ed. Boca Raton, Florida: Chapman; Hall/CRC. http://yihui.org/knitr/.

Xie, Yihui. 2022. Bookdown: Authoring Books and Technical Documents with R Markdown. https://CRAN.R-project.org/package=bookdown.