Sequential drug decision problems in long-term medical conditions: a case Study of Primary Hypertension Eunju Kim ba, ma, msc

4.4.4Transition probability

The SDDP of hypertension follows a semi-Markovian assumption during the drug switching period. The transition probability to the next state s_t+1 depends on SBP change and the risk of CVD, DM, other AEs and death, which are time-dependent. The transition probability is also modified depending on which treatment drug is currently in use.

In contrast, the long-term CVD model is a standard Markov model including Well, UA, MI, Stroke, HF, DM and Death. During the drug switching period, a proportion of patients in both Success and Failure, who have CVD or DM, move to the long-term CVD model. After the drug switching periods, all alive patients move to the long-term CVD model. 23 different sub-population groups are defined depending on when they move to the long-term CVD model. Each population has their own baseline risk of CVDs and DM based on the mean SBP and the variance (i.e., SD) when they enter the long-term CVD model. The conventional RR approach is used to consider the treatment effectiveness.