Two types of simulation can be used in a model: cohort or patient-level simulation. In the latter, each patient has a set of characteristics and a number of iterations is run to calculate the outcomes. The decision analysis usually follows a 5-step process: problem conceptualization; model conceptualization; model parameter estimation; run the model and interpretation; sensitivity analysis, transparency, and validation. A step that sometimes is undertaken before building a detailed decision tree is drawing an influence diagram containing decision and chance elements and outcomes. A particular model is generally used in case of diseases characterized by a gradual progression: Markov model. It considers several health states and transitions, to which probabilities are assigned. Time is divided into a series of sequential cycles; within each cycle, an individual must be in one state; transitions between the states occur at the end of each cycle. Partitioned survival models are characterized by a series of health states: The proportion of patients in each health state at each time point does not depend on transition probabilities, but is determined by a set of non-mutually exclusive survival curves. Discrete-event simulations are characterized by events that occur at an instant in time, resulting in a change of state in the system. The system is a chronological sequence of events. In the Discretely Integrated Condition Event (DICE) model, diseases can assume different levels over time and patients have different conditions varying over time. Events occurring at a particular time point can change the disease level or affect the occurrence of other events. The levels of conditions may change the probability of an event or its consequences. In agent-based models, individuals (agents) do not move between compartments, but change their internal state based on their interactions. Agents are characterized by activity, autonomy, and heterogeneity. Agents are active, while the environment (stage for agents' behaviors) is passive.