Meta-Analysis of Incidence Rate Data in the Presence of Zero-Event and Single-Arm Studies


  • Romain Piaget-Rossel University of Lausanne, Center for Primary Care and Public Health (Unisanté), Division of Biostatistics, Routedela Corniche 10,1010 Lausanne, Switzerland
  • Patrick Taffé University of Lausanne, Center for Primary Care and Public Health (Unisanté), Division of Biostatistics, Routedela Corniche 10,1010 Lausanne, Switzerland



Incidence rate, Meta-analysis, Negative-binomial model, Poisson model, Rare events, Random effects.


Unlike the classical two-stageDerSimonian and Laird meta-analysis method, the one-stage random-effectsPoisson and Negative-binomial models have the great advantage of including the information contained in studies reporting zero event in one or both arms and in studies with one missing arm. Since the Negative-binomial distribution relaxes the assumption of equi-dispersion made by the Poisson, it should perform better when data exhibit over-dispersion. However, the superiority of the Negative-binomial model with rare events and single-arm studies is unclear and needs to be investigated. Moreover, to the best of our knowledge, this model has never been investigatedin the context of a meta-analysis of incidence rate data with heterogeneous intervention effect. Therefore, we assessed the performance of the univariate and bivariate random-effects Poison and Negative-binomial models using simulations calibrated on a real dataset from a study onthe surgical management of phyllodes tumors. Results suggested that the bivariate random-effects Negative-binomial model should be favored for the meta-analysis of incidence rate data exhibiting over-dispersion, evenin the presence ofzero-event and single-arm studies.


Murad MH, Asi N, Alsawas M, Alahdab F. New evidence pyramid. BMJ Evid Based Med 2016; 21(4): 125-7. DOI:

Normand SLT. Meta-analysis: formulating, evaluating,

combining, and reporting. Stat Med 1999; 18: 321-59.<32 1::AID-SIM28>3.0.CO;2-P DOI:<321::AID-SIM28>3.0.CO;2-P

Nissen SE, Wolski K. Effect of Rosiglitazone on the risk of myocardial infarction and death from cardiovascular causes. N Engl J Med 2007; 356: 2457-71. DOI:

Niël-Weise BS, Stijnen T, van den Broek PJ. Anti-infective-treated central venous catheters: a systematic review of randomized controlled trials. Intensive Care Med 2007; 33: 2058-68. DOI:

DerSimonian R, Laird D. Meta-analysis in clinical trials. Control Clin Trials 1986; 7: 177-88. DOI:

Burke DL, Ensor J, Riley RD. Meta-analysis using individual participant data: one-stage and two-stage approaches, and why they may differ. Stat Med 2017; 36(5): 855-75. DOI:

Sweeting MJ, Sutton AJ, Lambert PC. What to add to nothing? Use and avoidance of continuity corrections in meta‐analysis of sparse data. Stat Med 2004; 23: 1351-75. DOI:

Friedrich JO, Adhikari NK and Beyene J. Inclusion of zero total event trials in meta-analyses maintains analytic consistency and incorporates all available data. BMC Med Res Methodol 2007; 7: 5. DOI:

Kuss O, Gummert JF and Borgermann J. Meta-analysis with rare events should use adequate methods. J Thorac Cardiovasc Surg 2008; 136(1): 241. DOI:

Keus F, Wetterslev J, Gluud C, et al. Robustness assessments are needed to reduce bias in meta-analyses that include zero-event randomized trials. Am J Gastroenterol 2009; 104(3): 546-51. DOI:

Piaget-Rossel R, Taffé P. Meta-analysis of rare events under the assumption of a homogeneous treatment effect. Biom J 2019; 1-18. DOI:

Piaget-Rossel R, Taffé P. A pseudo-likelihood approach for the meta-analysis of homogeneous treatment effects: exploiting the information contained in single-arm and double-zero studies. J Stat Adv Theory Appl 2019; 21: 91-117. 1365/images/JSATA7100122046RPiagetRossel.pdf DOI:

Kuss O. Statistical methods for meta-analyses including information from studies without any events – add nothing to nothing and succeed nevertheless. Stat Med 2015; 34: 1097-116. DOI:

Guevara JP, Berlin JA, Wolf FM. Meta-analytic methods for pooling rates when follow-up duration varies: a case study. BMC Med Res Methodol 2004; 4: 17. DOI:

Böhning D, Mylona K, Kimber A. Meta-analysis of clinical trials with rare events. Biom J 2015; 57: 633-48. DOI:

Spittal JM, Pirkis J, Gurrin CL. Meta-analysis of incidence rate data in the presence of zero events. BMC Med Res Methodol 2015; 15: 42. DOI:

Stijnen T. Hamza HT, Özdemir P. Random effects meta-analysis of event outcome in the framework of the generalized linear mixed model with applications in sparse data. Stat Med 2010; 29: 3046-67. DOI:

Hilbe JM. Negative binomial regression. Cambridge University Press; 2011. DOI:

Herbison P, Robertson MC, McKenzie JE. Do alternative methods for analysing count data produce similar estimates? Implications for meta-analyses. Syst Rev 2015; 4(1): 163. DOI:

Mullins P, Sharplin P, Yki-Jarvinen H, Riddle MC, Haring HU. Negative binomial meta-regression analysis of combined glycosylated hemoglobin and hypoglycemia outcomes across eleven Phase III and IV studies of insulin glargine compared with neutral protamine Hagedorn insulin in type 1 and type 2 diabetes mellitus. Clin Ther 2007; 29(8): 1607-19. DOI:

Toussaint A. ‘Width of margins in phyllodes tumors of the breast: the controversy drags on – a systematic review’. 2017. Diplôme interuniversitaire de sénologie et pathologie mammaire, François-Rabelais University and Rennes University, Tours and Rennes.

Agresti A, Hartzel J. Tutorial in biostatistics: Strategies for comparing treatment on a binary response with multi-center data. Stat Med 2000; 19: 1115-39.<11 15::AID-SIM408>3.0.CO;2-X DOI:<1115::AID-SIM408>3.0.CO;2-X

StataCorp. Stata Statistical Software: Release 15. 2017.College Station, TX: StataCorp LP.

Cai T, Parast L, Ryan L. Meta-analysis for rare events. Stat Med 2010; 29: 2078-89. DOI:

Lambert PC, Sutton AJ, Burton PR, Abrams KR, Jones DR. How vague is vague ? A simulation study of the impact of the use of vague prior distributions in MCMC using WinBUGS. Stat Med 2005; 24: 2401-28. DOI:

Senn S. Trying to be precise about vagueness. Stat Med 2007; 26: 1417-30. DOI:

Lambert D. Zero-inflated Poisson regression, with an application to defects in manufacturing. Technometrics 1992; 34: 1-14. DOI:

Böhning D, Kuhnert R, Rattanasiri S. Meta-analysis of binary data using profile likelihood. CRC Press; 2008. DOI:




How to Cite

Piaget-Rossel, R., & Taffé, P. (2019). Meta-Analysis of Incidence Rate Data in the Presence of Zero-Event and Single-Arm Studies. International Journal of Statistics in Medical Research, 8, 57–66.



General Articles