67, 068545/Z/02, 084762MA); and ZonMw: the Netherlands Organisation for Wellness Investigation and Improvement. Personal funding was provided by the Helsinki Biomedical Graduate System to D.L.C.; Netherlands Organization for Wellness Analysis and Improvement (grant numbers ZonMw 90700303, 916.10159) to V.W.V.J.; Dutch Kidney Foundation (grant number C08.2251) to H.R.T.; MRC Centre for Causal Analyses in Translational Epidemiology (grant number RD1634) to G.D.S., D.M.E. and N.J.T.; European Community’s Seventh Framework Programme (grant quantity FP7/2007-2013) and ENGAGE project, grant agreement HEALTH-F4-2007201413 to I.P. and V.L.; Tampere and Turku University Hospital Medical Funds (grant numbers 9M048, 9N035) to T.L.; Tampere Tuberculosis Foundation to T.L.; National Wellness and Healthcare Research Council (grant quantity 613608) to E.M.B. and Fellowship Scheme to G.W.M.; and Wellcome Trust Sir Henry Wellcome Postdoctoral Analysis Fellow (grant quantity 092447/Z/10/Z) to J.R.B.P. J.P.K. is funded by a Wellcome Trust 4-year PhD studentship in molecular, genetic, and life course epidemiology (WT083431MA).
Leroy et al. BMC Medical Research Methodology 2014, 14:17 http://biomedcentral/1471-2288/14/RESEARCH ARTICLEOpen AccessEstimating time-to-onset of adverse drug reactions from spontaneous reporting databasesFanny Leroy1,2*, Jean-Yves Dauxois3 , H e Th phile4,five , Fran ise Haramburu4,five and Pascale Tubert-Bitter1,Abstract Background: Analyzing time-to-onset of adverse drug reactions from remedy exposure contributes to meeting pharmacovigilance objectives, i.2-Chloro-4,6-dimethoxyaniline supplier e.1377584-27-4 uses identification and prevention.PMID:23600560 Post-marketing data are accessible from reporting systems. Times-to-onset from such databases are right-truncated since some individuals who were exposed towards the drug and who will at some point develop the adverse drug reaction may possibly do it just after the time of evaluation and hence are certainly not incorporated inside the information. Acknowledgment on the developments adapted to right-truncated information just isn’t widespread and these procedures have in no way been made use of in pharmacovigilance. We assess the use of appropriate strategies too because the consequences of not taking correct truncation into account (naive approach) on parametric maximum likelihood estimation of time-to-onset distribution. Approaches: Both approaches, naive or taking proper truncation into account, were compared having a simulation study. We made use of twelve scenarios for the exponential distribution and twenty-four for the Weibull and log-logistic distributions. These scenarios are defined by a set of parameters: the parameters on the time-to-onset distribution, the probability of this distribution falling inside an observable values interval as well as the sample size. An application to reported lymphoma just after anti TNF- treatment in the French pharmacovigilance is presented. Benefits: The simulation study shows that the bias as well as the imply squared error may in some situations be unacceptably big when suitable truncation will not be regarded whilst the truncation-based estimator shows normally much better and normally satisfactory performances and the gap may perhaps be massive. For the true dataset, the estimated anticipated time-to-onset results in a minimum distinction of 58 weeks among each approaches, that is not negligible. This difference is obtained for the Weibull model, below which the estimated probability of this distribution falling inside an observable values interval just isn’t far from 1. Conclusions: It is essential to take proper truncation into account for estimating time-to-onse.