Practical – cost-effectiveness of melioidosis vaccine:
Background
Melioidosis
Melioidosis is an infectious disease caused by bacteria Burkholderia pseudomallei.
The endemic areas are Northern Australia, Indian subcontinent, South America, and
Southeast Asia.(Figure 1)
Figure 1Worldwide distribution of melioidosis
Source: Cheng and Currie. 2005
Humans can be infected with B. pseudomallei through oral, nasal or skin exposure to
soil and water in endemic locations. Most patients are farmers and the poor from
rural areas. The incidence of infection is very high during rainy season when farmers
are most exposed to muddy environments. Melioidosis is considered a major cause
of community-acquired septicaemia in northeast Thailand (Chaowagul et al., 1989).
Underlying diseases
Common manifestations include cavitating pneumonia, hepatic and splenic
abscesses, and soft tissue and joint infections. Melioidosis has a very high in-hospital
mortality rate of up to 48%. Diabetes and renal impairment are common in the
northeast Thailand and both are significantly associated with melioidosis
(Chaowagul et al., 1993).
Burden of disease
According to the latest report from Limmathurotsakul (2010), the mortality rate of
severe melioidosis was 38% - 48% varied over 10 years records of Saprasitprasong
Hospital from 1997 – 2006 (Figure 2). Melioidosis incidence sharply increased from
year 2000 to 2006; 8.0 to 21.3 cases of melioidosis per 100,000 population in Ubon
Rachathani. The average mortality rate was 42.6% which made melioidosis the third
most common cause of death from infectious diseases after HIV AIDS and
tuberculosis in the Northeast. (Figure 3)
Figure 2 Incidence of melioidosis and associated death rate between 1997 and 2006
in Ubon Rachathani
Source: Limmathurotsakul et al., 2010
Figure 3 Mortality rates from infectious diseases per 100,000 people in Ubon
Ratchathani province between 1997 and 2006.
Source: Limmathurotsakul et al., 2010
Melioidosis is not only a life threatening disease, but also causes recurrent infections.
Recurrence can occur due to relapse (caused from the first infection) or reinfection
(new infection). Maharjan et al., 2005 reported that although patients had been treated
with proper antimicrobial treatment, melioidosis is still associated with a high
recurrence rate, with 75% relapsed and 25% experiencing reinfection. Recurrence time
varies in each patient from months to years. The median time of relapse was 228 days
and median time to reinfection was 823 days. The mortality rate of first relapse was
32% (Chaowagul et al., 1993).
Currently there is no vaccine for melioidosis although preliminary research is being carried out
towards the development of one. Whether or not such a vaccine would be cost effective is not
known. Your analysis should therefore consider the potential benefits of a melioidosis vaccine
in different target populations, while varying the estimates of its level and duration of protective
efficacy (PE) and its cost. The vaccine is assumed to reduce disease incidence, and mortality in
cases of severe illness, with an unknown duration of PE but a maximum value of 5 years. Health
benefits are measured in terms of quality adjusted life years (QALYs) gained [1]. The
incremental cost-effectiveness ratio (ICER) for a QALY gained should be compared initially with
the Thai GDP/capita which is a proxy measure for the assumed willingness to pay for a QALY
gained and tested against a higher value of up to 3 times DGP/capita [2, 3].
Part I. Deterministic Markov model
1. Open melioi TDmod.xls
2. The model parameters are given in the top left corner.
3. The transition probability matrix is given for the non-vaccine cohort. Replicate this with
the necessary adjustments in the vaccinated cohort.
4. The formula for the markov process are given for the non-vaccinated cohort; replicate
this in the vaccinated cohort.
5. Fill in the Results box for the vaccinated cohort.
6. Explore the impact of key parameters – which are most influential? Do a two way
sensitivity analysis for the cost of the vaccine, the duration of the PE:
a. In cell A42 enter the ICER (=D27)
b. Select the range A42 to F47
c. Go to the data tab and select What if, then data table
d. In the Row input cell enter the cell with the vaccine cost
e. In the Column input cell enter the cell with the PE
f.
Try creating a graph to show these results
Part II. Making the model probabilistic :
Open TDworkshop practical PSA.xlsm
Make sure macros are enabled
Add developer Tab to ribbon (Menu button in top left corner, select Popular and check the box
to show developer tab in the ribbon)
Fill in the probability distributions for PE mort and Duration using the same format as the cells
above.
Equate cells I4 and J4 to the relevant cells on the ‘Complete’ worksheet
Select the developer tab, choose ‘Insert’ and add a button from Form Control
Assign the macro ‘runmodel’ to this button (if not asked which macro to assign when inserting
the button, right click on it and choose relevant option).
Press Alt-F11 to access the VB editor and view the macro code.
Bo back to the spreadsheet and run the macro.
Create a scatterplot for the incremental effect on the X axis and the incremental cost on the Y
axis. Add a WTP threshold line (copy method from completed worksheet or try do this
yourself!)
In the VB editor, extend the selected cells to be copied from F4 to AI4.
Create a graph for the cost-effectiveness acceptability curve using the data from N1 to AI2.
Run the macro.
You’re done!