ISSN : 2146-3123
E-ISSN : 2146-3131

The Use of Cyclic Processes in Medical Decision Making: An Application of the Markov Model
Necdet Süt 1, Mevlüt Türe 2, Mustafa Şenocak 3
1Biostatistics Editor, Balkan Medical Journal Department of Biostatistics and Medical Informatics, Trakya University Faculty of Medicine, Edirne, Turkey
2Trakya Üniversitesi Tıp Fakültesi Biyoistatistik Anabilim Dalı, Edirne
3İstanbul Üniversitesi Cerrahpaşa Tıp Fakültesi Biyoistatistik Anabilim Dalı, İstanbul
Pages : 109-113

Abstract

Objectives: We aimed to explain the conceptual basis of the Markov model and to show the use of this model by an example application in medical decision making and medical predicting.

Study Design: An example model regarding the effectiveness of St. Jude Total Therapy XIIIB protocol in Acute Lymphoblastic Leukemia (ALL) was hypothesised to evaluate the Markov model concept. The expected remission probabilities in 10 cycles were calculated in a cohort simulation with 10,000 trials, in a cohort in remission in the initial state.

Results: Markov models are effective prediction models when the timing of events is important, when the decision problem involves risk over time and when events may happen more than once (as in recurrence). Markov models can be used in estimating such events. As a result of derived model, the remission probability without relaps of any case treatrd with St. Jude Total Therapy XIIIB protocol in ALL disease in the second cycle was found as 43% and it was sharply reduced after this cycle.

Conclusion: Cost, effectiveness, and health-related quality of life criteria of clinical strategies can be synthesised by the help of Markov models and used in the calculation of life expectancy, quality adjusted life expectancy and lifetime cost.

Keywords : Markov model; cyclic process; decision; prediction

Viewed : 2989
Downloaded : 922