Meta-analysis with zero-event studies: a comparative study with application to COVID-19 data

被引:29
作者
Wei, Jia-Jin [1 ]
Lin, En-Xuan [2 ]
Shi, Jian-Dong [1 ]
Yang, Ke [1 ]
Hu, Zong-Liang [3 ]
Zeng, Xian-Tao [4 ]
Tong, Tie-Jun [1 ]
机构
[1] Hong Kong Baptist Univ, Dept Math, Hong Kong, Peoples R China
[2] Shenzhen Res Inst Big Data, Shenzhen, Peoples R China
[3] Shenzhen Univ, Coll Math & Stat, Shenzhen, Peoples R China
[4] Wuhan Univ, Zhongnan Hosp, Ctr Evidence Based & Translat Med, Wuhan, Peoples R China
基金
中国国家自然科学基金;
关键词
Continuity correction; Coronavirus disease 2019 data; Meta-analysis; Relative risk; Zero-event studies; RANDOM-EFFECTS MODELS; BINARY OUTCOMES; LOGARITHM; RATIO; RISK; BIAS; ADD;
D O I
10.1186/s40779-021-00331-6
中图分类号
R5 [内科学];
学科分类号
1002 ; 100201 ;
摘要
Background Meta-analysis is a statistical method to synthesize evidence from a number of independent studies, including those from clinical studies with binary outcomes. In practice, when there are zero events in one or both groups, it may cause statistical problems in the subsequent analysis. Methods In this paper, by considering the relative risk as the effect size, we conduct a comparative study that consists of four continuity correction methods and another state-of-the-art method without the continuity correction, namely the generalized linear mixed models (GLMMs). To further advance the literature, we also introduce a new method of the continuity correction for estimating the relative risk. Results From the simulation studies, the new method performs well in terms of mean squared error when there are few studies. In contrast, the generalized linear mixed model performs the best when the number of studies is large. In addition, by reanalyzing recent coronavirus disease 2019 (COVID-19) data, it is evident that the double-zero-event studies impact the estimate of the mean effect size. Conclusions We recommend the new method to handle the zero-event studies when there are few studies in a meta-analysis, or instead use the GLMM when the number of studies is large. The double-zero-event studies may be informative, and so we suggest not excluding them.
引用
收藏
页数:11
相关论文
共 47 条
[1]  
Agresti A., 2012, CATEGORICAL DATA ANA
[2]   Meta-analysis of binary outcomes via generalized linear mixed models: a simulation study [J].
Bakbergenuly, Ilyas ;
Kulinskaya, Elena .
BMC MEDICAL RESEARCH METHODOLOGY, 2018, 18
[3]   How to perform a meta-analysis with R: a practical tutorial [J].
Balduzzi, Sara ;
Ruecker, Gerta ;
Schwarzer, Guido .
EVIDENCE-BASED MENTAL HEALTH, 2019, 22 (04) :153-160
[4]   Methods for evidence synthesis in the case of very few studies [J].
Bender, Ralf ;
Friede, Tim ;
Koch, Armin ;
Kuss, Oliver ;
Schlattmann, Peter ;
Schwarzer, Guido ;
Skipka, Guido .
RESEARCH SYNTHESIS METHODS, 2018, 9 (03) :382-392
[5]  
Borenstein M., 2021, Introduction to meta-analysis
[6]   A basic introduction to fixed-effect and random-effects models for meta-analysis [J].
Borenstein, Michael ;
Hedges, Larry V. ;
Higgins, Julian P. T. ;
Rothstein, Hannah R. .
RESEARCH SYNTHESIS METHODS, 2010, 1 (02) :97-111
[7]  
Carter RE, 2010, J R STAT SOC C-APPL, V59, P657
[8]   Meta-analysis of studies with bivariate binary outcomes: a marginal beta-binomial model approach [J].
Chen, Yong ;
Hong, Chuan ;
Ning, Yang ;
Su, Xiao .
STATISTICS IN MEDICINE, 2016, 35 (01) :21-40
[9]   An Empirical Bayes Method for Multivariate Meta-analysis with an Application in Clinical Trials [J].
Chen, Yong ;
Luo, Sheng ;
Chu, Haitao ;
Su, Xiao ;
Nie, Lei .
COMMUNICATIONS IN STATISTICS-THEORY AND METHODS, 2014, 43 (16) :3536-3551
[10]   Physical distancing, face masks, and eye protection to prevent person-to-person transmission of SARS-CoV-2 and COVID-19: a systematic review and meta-analysis [J].
Chu, Derek K. ;
Akl, Elie A. ;
Duda, Stephanie ;
Solo, Karla ;
Yaacoub, Sally ;
Schunemann, Holger J. .
LANCET, 2020, 395 (10242) :1973-1987