A Pointwise Approach to Dose-Response Meta-Analysis of Aggregated Data


  • Alessio Crippa Department of Public Health Sciences, Karolinska Institutet, Sweden
  • Ilias Thomas Department of Micro-Data Analysis, Dalarna University, Sweden
  • Nicola Orsini Department of Public Health Sciences, Karolinska Institutet, Sweden




Dose-response, Meta-analysis, Pointwise average, Flexible model.


In a two-stage dose-response meta-analysis a common functional relationship is applied to each study and an overall curve is obtained by combining study-specific dose-response coefficients. Possible limitations are: 1) a common dose-response model may have a poor fit in some of the studies; 2) combining dose-response coefficients discard information about study-specific exposure range. A pointwise approach for meta-analysis may overcome those limitations by combining predicted relative risks for a fine grid of exposure values based on potentially different dose-response models.

We described how to flexibly model the dose-response association in a single study using fractional polynomials and spline, and how to present the combined results from study-specific analyses.

The strategy is illustrated using aggregated data derived from the Surveillance, Epidemiology, and End Results program, with results compared to the corresponding analysis based on individual data.

Another example on milk consumption and all-cause mortality is used to show the advantages of the pointwise approach regarding flexibility in the dose-response analyses, limitations of extrapolations, and informativeness in presenting pooled results.

Application of the proposed strategy may improve dose-response meta-analysis of observational studies in case of particularly heterogeneous exposure distributions.


Orsini N, Li R, Wolk A, Khudyakov P, Spiegelman D. Meta-Analysis for Linear and Nonlinear Dose-Response Relations: Examples, an Evaluation of Approximations, and Software. Am J Epidemiol 2012; 175(1): 66-73. https://doi.org/10.1093/aje/kwr265 DOI: https://doi.org/10.1093/aje/kwr265

Multivariate Dose-Response Meta-Analysis: The dosresmeta R Package | Crippa | Journal of Statistical Software. [Online]. Available: https://www.jstatsoft.org/article/view/v072c01/0. [Accessed: 23-Jan-2017].

Discacciati A, Crippa A, Orsini N. Goodness of fit tools for dose-response meta-analysis of binary outcomes. Res Synth Methods 2015. DOI: https://doi.org/10.1002/jrsm.1194

Crippa A, Discacciati A, Bottai M, Spiegelman D, Orsini N. One-stage dose–response meta-analysis for aggregated data. Stat Methods Med Res In press, pp. 1-18.

Gasparrini A, Armstrong B, Kenward MG. Multivariate meta-analysis for non-linear and other multi-parameter associations. Stat Med 2012; 31(29): 3821-3839. https://doi.org/10.1002/sim.5471 DOI: https://doi.org/10.1002/sim.5471

Rota M, et al. Random-effects meta-regression models for studying nonlinear dose–response relationship, with an application to alcohol and esophageal squamous cell carcinoma. Stat Med 2010; 29(26): 2679-2687. https://doi.org/10.1002/sim.4041 DOI: https://doi.org/10.1002/sim.4041

Bagnardi V, Zambon A, Quatto P, Corrao G. Flexible Meta-Regression Functions for Modeling Aggregate Dose-Response Data, with an Application to Alcohol and Mortality. Am J Epidemiol 2004; 159(11): 1077-1086. https://doi.org/10.1093/aje/kwh142 DOI: https://doi.org/10.1093/aje/kwh142

Sauerbrei W, Royston P. A new strategy for meta-analysis of continuous covariates in observational studies. Stat Med 2011; 30(28): 3341-3360. https://doi.org/10.1002/sim.4333 DOI: https://doi.org/10.1002/sim.4333

Greenland S, Longnecker MP. Methods for trend estimation from summarized dose-response data, with applications to meta-analysis. Am J Epidemiol 1992; 135(11): 1301-1309. https://doi.org/10.1093/oxfordjournals.aje.a116237 DOI: https://doi.org/10.1093/oxfordjournals.aje.a116237

Royston P. A strategy for modelling the effect of a continuous covariate in medicine and epidemiology. Stat Med 2000; 19(14): 1831-1847. https://doi.org/10.1002/1097-0258(20000730)19:14<1831::AID-SIM502>3.0.CO;2-1 DOI: https://doi.org/10.1002/1097-0258(20000730)19:14<1831::AID-SIM502>3.0.CO;2-1

Harrell FE. Regression Modeling Strategies. New York, NY: Springer New York, 2001. https://doi.org/10.1007/978-1-4757-3462-1 DOI: https://doi.org/10.1007/978-1-4757-3462-1

DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials 1986; 7(3): 177-188. https://doi.org/10.1016/0197-2456(86)90046-2 DOI: https://doi.org/10.1016/0197-2456(86)90046-2

Tai P, et al. Modeling the effect of age in T1-2 breast cancer using the SEER database. BMC Cancer 2005; 5: 130. https://doi.org/10.1186/1471-2407-5-130 DOI: https://doi.org/10.1186/1471-2407-5-130

Larsson SC, Crippa A, Orsini N, Wolk A, Michaëlsson K. Milk Consumption and Mortality from All Causes, Cardiovascular Disease, and Cancer: A Systematic Review and Meta-Analysis. Nutrients 2015; 7(9): 7749-7763. https://doi.org/10.3390/nu7095363 DOI: https://doi.org/10.3390/nu7095363

Debray TPA, et al. Get real in individual participant data (IPD) meta-analysis: a review of the methodology. Res Synth Methods 2015; 6(4) 293-309. https://doi.org/10.1002/jrsm.1160 DOI: https://doi.org/10.1002/jrsm.1160

Sutton AJ, Kendrick D, Coupland CAC. Meta-analysis of individual- and aggregate-level data. Stat Med 2008; 27(5): 651-669. https://doi.org/10.1002/sim.2916 DOI: https://doi.org/10.1002/sim.2916

Stewart LA, Tierney JF. To IPD or not to IPD? Advantages and Disadvantages of Systematic Reviews Using Individual Patient Data. Eval Health Prof 2002; 25(1): 76-97. https://doi.org/10.1177/0163278702025001006 DOI: https://doi.org/10.1177/0163278702025001006

Colditz GA, Burdick E, Mosteller F. Heterogeneity in meta-analysis of data from epidemiologic studies: a commentary. Am J Epidemiol 1995; 142(4): 371-382. https://doi.org/10.1093/oxfordjournals.aje.a117644 DOI: https://doi.org/10.1093/oxfordjournals.aje.a117644

Blettner M, Sauerbrei W, Schlehofer B, Scheuchenpflug T, Friedenreich C. Traditional reviews, meta-analyses and pooled analyses in epidemiology. Int J Epidemiol 1999; 28(1): 1-9. https://doi.org/10.1093/ije/28.1.1 DOI: https://doi.org/10.1093/ije/28.1.1




How to Cite

Crippa, A., Thomas, I., & Orsini, N. (2018). A Pointwise Approach to Dose-Response Meta-Analysis of Aggregated Data. International Journal of Statistics in Medical Research, 7(2), 25–32. https://doi.org/10.6000/1929-6029.2018.07.02.1



General Articles