Local EM estimation of the hazard function for interval-censored data

We propose a smooth hazard estimator for interval-censored survival data us ing the method of local likelihood. The model is fit using a local EM algor ithm. The estimator is more descriptive than traditional empirical estimate s in regions of concentrated information and takes on a parametric flavor i n regions of sparse information. We derive two different standard error est imates for the smooth curve, one based on asymptotic theory and the other o n the bootstrap. We illustrate the local EM method for times to breast cosm esis deterioration (Finkelstein, 1986, Biometrics 42, 845-854) and for time s to HIV-1 infection for individuals with hemophilia (Kroner et al., 1994, Journal of AIDS 7, 279-286). Our hazard estimates for each of these data se ts show interesting structures that would not be found using a standard par ametric hazard model or empirical survivorship estimates.