The treatment of patients affected by multiple diseases (comorbid patients) is one of the main challenges of the modern healthcare, involving the analysis of the interactions of the guidelines for the specific diseases. Practically speaking, such interactions occur in time. The GLARE project explicitly provides temporal representation and temporal reasoning methodologies to cope with such a fundamental issue. In this paper, we propose a further improvement, to take into account that, often, (i) the actions in the guidelines can be executed by physicians at different times with different preferences, and that (ii) the effects of such actions have a probabilistic distribution in time. In our approach, physicians may investigate what are the preferences of their choices on the execution-time of guideline actions, and the probabilities that their effects temporally intersect (interactions may occur only in case effects intersect in time).