Fig. 3From: Bias due to censoring of deaths when calculating extra length of stay for patients acquiring a hospital infectionEstimated Cumulative hazards rates in the first 80 days for the multi-state models in Figs. 1 and 2. The slope of each line corresponds to the actual hazard rate, e.g a straight line would mean a constant hazard rate. The left figure shows the cumulative hazard functions for model A, when death is considered as competing event. The right figure corresponds to that of model B, when the patients are censored at the time of deathBack to article page