Flexible piecewise linear model for investigating dose-response relationship in meta-analysis: methodology, examples, and comparison

, , , , ,
1 Chinese Evidence based medicine Center, West China Hospital, Chengdu, China
2 Department of Health Research Methods, Evidence, and Impact, McMaster University, Hamilton, Canada
3 Department of Urology, Zhongnan Hospital, Wuhan University, Wuhan, China
DOI
10.7287/peerj.preprints.27277v1
Published
Accepted
Subject Areas
Epidemiology, Evidence Based Medicine, Internal Medicine, Public Health, Statistics
Keywords
piecewise linear function, dose-response meta-analysis, discrete exposure
Copyright
© 2018 Xu et al.
Licence
This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. For attribution, the original author(s), title, publication source (PeerJ Preprints) and either DOI or URL of the article must be cited.
Cite this article
Xu C, Thabane L, Liu T, Li L, Borhan S, Sun X. 2018. Flexible piecewise linear model for investigating dose-response relationship in meta-analysis: methodology, examples, and comparison. PeerJ Preprints 6:e27277v1

Abstract

Objectives: Dose-response meta-analysis (DRMA) is widely employed to establishing the potential dose-response relationship between continuous exposures and disease outcomes. However, no method is readily available for exploring the relation between a discrete exposure and a binary or continuous outcome. We proposed a piecewise linear (PL) DRMA model as a solution to this issue.

Methods: We illustrated the methodology of PL model in both one-stage DRMA approach and two-stage DRMA approach. The method by testing the equality of slopes of each piecewise was employed to judge if there is “piecewise effect” against simple linear trend. We then used sleep (continuous exposure) and parity (discrete exposure) data as examples to illustrate how to apply PL model in DRMA using the Stata code attached. We also empirically compared the slopes of PL model with simple linear as well as restricted cubic spline (RCS) model.

Results: Both one-stage and two-stage PL DRMA model fitted well in our examples, and the results were similar. Obvious “piecewise effects” were detected in both the two examples by the method we used. In our example, the PL model showed better fitting effect and practical reliable results compared to simple linear model, while similar results for to RCS model.

Conclusion: Piecewise linear function is a simple and valid method for DRMA and can be used for discrete exposures. It also represents a superior model to linear model in DRMA and may be an alternative model to non-linear model.

Author Comment

This is a submission to PeerJ for review.

Supplemental Information

Table S1 and Table S2

Table S1. The raw data for sleep duration and risk of all cause mortality

Table S2. The raw data for parity and risk of rheumatoid arthritis.

DOI: 10.7287/peerj.preprints.27277v1/supp-1