Investigating the utility of multivariate meta-analysis methods in estimating summary dose response curve

Deepthy Melepurakkal Sadanandan, Kalesh M Karun, Harichandrakumar K T, N. Sreekumaran Nair


Background: Traditional meta-analyses often assess the effectiveness of different doses of the same intervention separately or examine overall differences between the intervention and placebo. The study aimed to model the effect sizes obtained for different doses from multiple studies through a two stage dose response meta-analytic approach while taking dose variations into account.

Materials and Methods: Different dose response meta-analysis models using linear, quadratic and restricted cubic spline functions were fitted. A two stage approach utilizing the multivariate meta-analysis was performed and results obtained were compared with univariate meta-analysis. Random effect dose response meta-analysis was performed by using data from an existing systematic review on combination therapy with Zonisamide and anti-Parkinson’s drugs for Parkinson’s disease. The effective dose or optimum dose to produce maximum response was also being investigated.

Results: Dose response meta-analysis was performed using the data from four double blinded randomised controlled trials with 724 and 309 Parkinson’s patients in dose arms placebo arms. The quadratic model gave the smallest AIC when compared to linear and restricted cubic spline models indicating best fit to the data.

Conclusion: Compared to the traditional approach, the two stage approach could model the dose dependent effect of Zonisamide on UPDRS part III score and predict the outcome for different doses through a single analysis.


Dose-response meta-analysis; Restricted Cubic Spline model; Multivariate meta-analysis; Parkinson’s disease; Zonisamide

Full Text: PDF

JRHS Office:

School of Public Health, Hamadan University of Medical Sciences, Shaheed Fahmideh Ave. Hamadan, Islamic Republic of Iran

Postal code: 6517838695, PO box: 65175-4171

Tel: +98 81 38380292, Fax: +98 81 38380509