by John P. A. Ioannidis, Lamberto Manzoli, Corrado De Vito, Maddalena D'Addario, Paolo Villari
Randomized evidence for vaccine immunogenicity and safety is urgently needed in the setting of pandemics with new emerging infectious agents. We carried out an observational survey to evaluate how many randomized controlled trials testing 2009 H1N1 vaccines were published among those registered, and what was the time lag from their start to publication and from their completion to publication. Methods
PubMed, EMBASE and 9 clinical trial registries were searched for eligible randomized controlled trials. The units of the analysis were single randomized trials on any individual receiving influenza vaccines in any setting. Results
73 eligible trials were identified that had been registered in 2009–2010. By June 30, 2011 only 21 (29%) of these trials had been published, representing 38% of the randomized sample size (19905 of 52765). Trials starting later were published less rapidly (hazard ratio 0.42 per month; 95% Confidence Interval: 0.27 to 0.64; p<0.001). Similarly, trials completed later were published less rapidly (hazard ratio 0.43 per month; 95% CI: 0.27 to 0.67; p<0.001). Randomized controlled trials were completed promptly (median, 5 months from start to completion), but only a minority were subsequently published. Conclusions
Most registered randomized trials on vaccines for the H1N1 pandemic are not published in the peer-reviewed literature.