Dear list, I am wondering whether R allows to perform a Poisson regression on counting data which include censored observations, that is, observations of the form "2 events or more" or "less than 2 events" or even "1 or 2 events". Can anyone give me a hint whether this is already possible and how to do it? Data of the form (obstime is the observation time, "offset" in Poisson regression): obs obstime cmin cmax 1 1 2 Inf 2 2 0 1 3 1 1 2 and so on. Best wishes, Justine --