According to the graph, uransition probability initialed with"p" iollowed by

the first letter of states. For example: pStoD is for transition probability from
stroke state to death state. Transi- tion probabilities(represented by arrows
are derived from clinical trials, observational studies or from systematic
reviews. Transition probabilities can be structured by transition probability
matrix. on Table 2. The"loop" symbol means that patients may remain
experiencing with health condition in that state at specific time before
moving to another state. In CUA sometimes an ideal model is added with
second year ofprogressive states such as:poststroke, post MI, second year
lung cancer and soon, with an assumption that progressivity of utility values
occur in patients after receiving the intervention. If the model having
extended structure such as post-stroke or post MI. it means that we could
add new input parameters for second year onwards.

The total sum of transition probabilities from initial until end ofmodel must
be 1.0. There fore, when we calc ate the probability of patients who renain
in the same states(loop sym with probabilties. Value T" imply representing
that there is no transiticn probability fron a state to anoth stare(see figure
Sometimes, when it is intended to input parameters such as transition
probabilities, for in stancc from systematic reviews, the confusion
bctwccn"ratc" and"probability" cxists.Transition probabilities gathered from
literature reviews do not exactly renect the model's cycle that we con
struct(Sato 201U) luatcs arc potentual occurrence of an event in dchined
population and dctincd time. Meanwhile, probability is assigned values
ranging from 0 and 1, representing the likelihood ot an event happening
over a spoc tic pcriod ot timc". Lo dcal with th s, rate 13 possible to bc con
verted to a probability and vice versa(Briggs et al. 2006) where p is the
probability, r is the rate and t is thc timc pcriod of an intercst: p-i-exp-rt In
thc cohort hypothetical simulation, thc author dcsribcs bow Markov model is
running, without assigning all input parameters. Details are provided in
appendix l. We only incorporated transition probability for every health
states. anly for familiarizing audience afbasic Markov ap plication. Unlike a
decision tree, Markov model is applicable for applying discounting, which is
an important uur in health euoruumic evaluatiun Additionally, iar chronic
diseases, the lie lable and survival dala ars ofen aluehled into the mudel
Quality Adjusted Life Years(QALYs) also can he estimated as final results,
sinne in calo lating QAI ng quality of life enrompasses thc length of timo
spent in health statcs(Briggs, 199H). The cost analysis also has similar
mcthod. costs arc rcprcscntcd in treatment events. With computcr
application, this complex sumulatan can be handled. After simulating
decision enalytic model, the final results are ICER. derived from dimerences
for both costs and eTects utility, in this case after 20 years cycle, Thus,
dacision rules should be followed.

Although Nhai Kov model application is neiprul in decision analytic modeling,

there several limitations that we should aware of Markov has a characteristic
of memory less" (Son enberg, 199 Briggs, 19UN). It means when patient
move to another states, the model will have no historical memory where
patients has come from previous cyc The population equal in this model also
with constant risk. Adding additional states and time dependency into
transition prob ahility may poss ble for the model lesides, Markov model has
greater complexity compared to decision tree, however computer software
nowadays could help for running this model Ilandling uncertainty Variability.
uncertainty and heterogeneity the results influences of decision analytic
model, hese should he handled 4) The results of evaluation are depending
on mncertain arton et a es that related with various factors. This section will
focus on short-review for handling uncer tainty, before going further, that
will be useful to define several types of uncertainty in modeling Briggs et al,
2012) Decision makers needs evidence beyond point estimate of the
outcome, The uncertainty surrounding ouwcome should be explored and
repurted. Uncertainty may cateuorized as methodological uncertainty.
parameter unccrtainty, structural uncertainty and gencralizability Briggs and
Sclupl ner, 1995). Thus, decisio maker can receive more info rmation what
factors that influence the model and ts findings, and increasing the
confidence level where they read and intend to acccpt the cvaluation results
Uncertainty can be handled by performing appropriate sensitivity analysis.
There are sev eral types of sensitivity analysis(Andronis et al. 2009, Briggs
et al. 2012). First, a one-way sen sitivity analysis It the sim est form of
sensitivity analysis, exam nes impact of the changes in model results by
varying plausible values range. Tornado diagram is used to visualized this
rescated with basclince value and range variation in the right and left wing
Sccond is multi-way sensitivity analysis, including two-way sensitivity
analysis. This examines relationship of two or more parameters, not single
parameters like one-way sensitivity analyses. Another type is extreme
scenario. simply when thc paramctcrs or input value of favorablc
intervention anc cstablished to rovide best and worst case scenario
Probabilistic Sensitivity Analysis(PSA) is one type of sensitivity analysis that
runs ran domly by large namber of Monte Carlo simulation(for instance: 10,
IXX) iterative) t 15 examines each input parameters with their probability
distributions. lt resulted mean cost as well as effec iveness. This approach is
preferrcd to generatc cost-cffoctivcncss acccptability curvc(Claxton c al.
2005, Ades et al. 2006) Finally, model result and its sensitivity analysis
should be presented clearly. It is advisable to sce guidelines rclated to full
hcalth cconomics study reporting(Huscrcau ct al. 2013 Conclusions Models
are chosen, construct and developed based on decision problems and must
be ap propriate with decision makers purpose. All evidence as input
parameters are assigned and per forms by mathematical operations that
translating into model. Furthermore, to decide the set of alterna that should
be chosen, the final results of decision analytical models have to follow
decision Tules for cost-effectiveness of heal h intervention. The ex this paper
not pro- anation viding best practice yet simple example of modeling
application, further details and intemational puidelines are availah e
elsewhere(Schipher et al. 2 Ilyelmegren et al. 2 0001: Weinstein et al.

2003; Philips et al. 20U6) This papcr providcs a revicw and application of
dccision analytic modcl in cconomic cval- uation, with tobacco cessation
examples. Decision tree and Markov model are commonly per Turuned in
heal ecunomic evaluation. deed willi their advantages and liinilauious.
These Lwo model are complements each other. Performing decision analytic
modeling is beneficial, specifi- cally when there arc scvcral barricrs from
clinical trials. Decision trce with its simplicity is ablc to provide clear
pathways of decision and for economic evaluation with intermediate
outcomes. Markov tueanwhile has abilily Lo analyze uie condition with
longer time orizon and expecting final outcomes Acknowledgements The
author thanks uu Prof. Hasbullalu Tliabrany MD.,MPH. DrPH ior his valuable
inpul and review on tinal draft of manuscript.