Objective Evaluating the efficacy of Shufeng Piangchan Tang combined with conventional western medicine in the treatment of Parkinson's disease in the elderly based on Markov (Markov) modeling. MethodsA total of 220 elderly PD patients admitted to the hospital from January 2018 to May 2021 were continuously selected and divided into western medicine treatment group and integrated Chinese and Western medicine treatment group according to random number table method. Parkinson's Scale (UPDRS) scores and serum levels were recorded before and after treatment. Based on the real follow-up data, Markov model was built and verified.Results After treatment, the UPDRS score of the combined Chinese and western medicine treatment group was lower than that of the western medicine treatment group (P<0.05), and the levels of BDNF, EGF and SOD were higher than those of the Western medicine treatment group (P<0.05). The transfer frequency of UPDRS score in the Western medicine treatment group and the combined Chinese and Western medicine treatment group 6 months after treatment was taken as the transfer probability of Markov model, the change of UPDRS score status in the Western medicine treatment group and the combined Chinese and Western medicine treatment group 6 months, 1 year and 2 years was predicted. The results showed that compared with the baseline UPDRS score, the proportion of good UPDRS score and good UPDRS score in the two groups increased significantly, and the proportion of good UPDRS score and good UPDRS score in the Chinese and western treatment group was higher than that in the western treatment group at different time periods (P<0.05). The 6-month, 1-year and 2-year health utility values (0.182, 0.194, 0.089) in the integrated Chinese and western medicine treatment group were higher than those in the Western medicine treatment group (0.127, 0.114, 0.052). Follow-up 2 years after treatment, the actual UPDRS score status distribution of different treatment groups was obtained, which was compared with the predicted UPDRS score status distribution obtained by Markov model. There was no significant difference between the predicted frequency and the actual frequency of UPDRS score distribution in the Western medicine treatment group and the combined treatment group (P>0.05), indicating a high degree of model fitting. Conclusion Based on the results predicted by Markov model, the combined treatment of traditional Chinese and Western medicine has better long-term clinical efficacy than that of Western medicine alone in the treatment of elderly PD