cddep modelingscenarios 20120915 0
DESCRIPTION
http://www.cddep.org/sites/default/files/cddep-modelingscenarios_20120915_0.pdfTRANSCRIPT
Modeling)Future)AMFm)Scenarios)
AMFm)Scenarios)Moving)Forward)• Model&Structure&• Inputs&• Scenarios&
• Child4Targe7ng&• Par7al&Subsidy&
• Summary&
2&
Model)Structure)
3&
Fever&
Transmission&
Infec7on&
Hospitaliza7on&
Death&
Rx&
Rx&
Rx&An7malarial&Drugs&Avert&Bad&Outcomes,&&&the&Timing&of&Treatment&MaKers&
Model)Structure)
4&Time (in Years)
Dea
ths
per 1
00,0
00, p
er y
ear
Dos
es p
er 1
00,0
00, p
er y
ear
Deaths Averted
Baseline
T0
0
Introduction of a New Drug
Model)Structure)
5&
Time (in Years)
Dea
ths
per 1
00,0
00, p
er y
ear
Dos
es p
er 1
00,0
00, p
er y
ear
Freq
uenc
y of
Res
ista
nce
01
Deaths Averted
Baseline
T0
0
Intro
duct
ion
of a
New
Dru
g
● Emergence
Model)Structure)
6&
Time (in Years)
Intro
duct
ion
of A
CTs
Deaths Averted
Dea
ths
per 1
00,0
00, p
er y
ear
Dos
es p
er 1
00,0
00, p
er y
ear
Freq
uenc
y of
Res
ista
nce
01
●
Model)Structure)
7&
annual EIR
annu
al F
OI
a
0.01 0.1 1 10 100 1000
1/10
01
10
●●
●●●● ● ● ●
●●
●● ●
●
annual EIR
Tran
smiss
ion
Effic
ienc
y b
0.01 0.1 1 10 100 1000
1/50
01/
103/
5
●
●
●●●
●
●
●
●
●
●
●●
● ●
Malaria&Transmission&by&Mosquitoes&
Model)Structure)
8&
S& I&
S& I&
Children&
Adults&
• Two4stage&model&of&malaria&transmission&• Children&are&more&likely&to&become&clinically&ill&when&infected&• Treated&individuals&remain&uninfected&un7l&drug&clears&system&
P&
P&
Model)Structure)
9&
S&
IW&
IS&
ID&
IT&
Drug&sensi7ve&
Drug&Resistant&(Single)&NAD,&AMT,&PMT&
Drug&Resistant&(Double)&NAD4PMT,&NAD4AMT,&ACT&
Drug&Resistant&(All)&NAD4ACT&
Individuals&(both&children&and&adults)&can&be&infected&by&drug&sensi7ve&or&drug&resistant¶sites&
NAD&4&Non4Artemisinin&Drugs&AMT&4&Artemisinin&Monotherapy&PMT&4&Partner&Drug&Monotherapy&
Inputs)
10&Source:&Murray,&C.&J.&L.,&L.&C.&Rosenfeld,&et&al.&(2012).&"Global&malaria&mortality&between&1980&and&2010:&a&systema7c&analysis."&The$Lancet&379(9814):&4134431&
Death&Rates&
Inputs)
11&Source:&CHAI&evalua7on&of&DHS&Household&Surveys&
Fever&Rates& Treatment&Rates&
Price&
Ini7al&Market&Share&
Inputs:)Demand)Functions)
12&
Increasing&the&subsidy&increases&the&demand&for&ACTs&as&well&as&overall&drug&demand& 0 1 2 3 4
0
20
40
60
80
100
Ghana
Price Difference of Subsidy
Dru
g P
erce
ntag
es
CQAMT
ACT-qACT-n
0 1 2 3 4
0.000
0.005
0.010
0.015
Ghana
Price Difference of Subsidy
Dru
g D
eman
d
Non-ARTAMT
ACT-qACT-n
all
13&
0.0 0.2 0.4 0.6 0.8 1.0
020
4060
8010
0Kenya
Price Difference of Subsidy
Dru
g P
erce
ntag
es
CQAMT
ACT-qACT-n
0 1 2 3 4 5 6
020
4060
8010
0
Madagascar
Price Difference of Subsidy
Dru
g P
erce
ntag
es
CQAMT
ACT-qACT-n
0.0 0.5 1.0 1.5
020
4060
8010
0
Niger
Price Difference of Subsidy
Dru
g P
erce
ntag
es
CQAMT
ACT-qACT-n
0 1 2 3 4
020
4060
8010
0
Nigeria
Price Difference of Subsidy
Dru
g P
erce
ntag
es
CQAMT
ACT-qACT-n
0 1 2 3 4
020
4060
8010
0
Tanzania
Price Difference of Subsidy
Dru
g P
erce
ntag
es
CQAMT
ACT-qACT-n
0.0 0.2 0.4 0.6 0.8
020
4060
8010
0
Uganda
Price Difference of SubsidyD
rug
Per
cent
ages
CQAMT
ACT-qACT-n
Scenarios)• Child4Targe7ng&
• Subsidize&child&packets&only&• Assume&different&levels&of&leakage&to&adults&
• Adults&that&take&child&packs&either&“stack”&or&underdose&• Par7al&Subsidy&
• Pricing&Op7ons&• No4subsidy &&• Par7al&Subsidy&• Full&Subsidy&
• Tiering&Op7ons&
14&
Scenarios:)Baseline)
15&
Annual&ACT&Treatments&(high&and&low&elas7city)&
• Universal&Subsidy&• Assumes&a&propor7on&
with&no&malaria&infec7ons&
• Individuals&treat&• Child&doses&cost&50%&
of&adults&(~$1)&
Number&of&es7mated&annual&ACT&treatments&demanded&
Scenarios:)Baseline)
16&
Number&of&Deaths&Averted&(high&and&low&elas7city)&
Scenarios:)Baseline)
17&
Number&of&Deaths&Averted&(high&and&low&elas7city)&
Same&as&prior&slide&–&different&scale&
Scenarios:)Child=Targeted)
18&
Number&of&Deaths&Averted&(low&elas7city)&
Scenarios:)Child=Targeted)
19&
Number&of&Deaths&Averted&(low&elas7city)&
Same&as&prior&slide&–&different&scale&
Scenarios:)Child=Targeted)
20&
Number&of&ACT&Treatments&Demanded&(low&elas7city)&
Scenarios:)Child=Targeted)
21&
Cost4Effec7veness&(low&elas7city)&
Partial)Subsidy)• No&Subsidy&
• Countries&have&access&to&AMFm&mechanism&to&purchase&qa4ACTs&at&nego7ated&price&(but&no&subsidy)&
• Reduces&end4user&price&• Par7al&Subsidy&
• Subsidy&covers&a&por7on&(50%)&of&the&cost&of&quality4assured&ACTs,&but¬&the&full&price.&&
• Results&in&larger&reduc7on&in&end4user&price&• Full&Subsidy&
• Subsidy&covers&95%&of&the&cost&of&quality4assured&ACTs&• Results&in&significant&reduc7on&in&end4user&price&
22&
Partial)Subsidy)Cost)by)Country)
23&
Low&Elas7city4Low&Ini7al&ACT&Use&
Annu
al&Sub
sidy&Co
st&($
millions)&
Partial)Subsidy)Cost)
24&Low&Ini7al&ACT&Use&
Low&Ini7al&ACT&Use&
Higher&Ini7al&ACT&Use&
Higher&Ini7al&ACT&Use&
Low&Elas7city& High&Elas7city&
Annu
al&Sub
sidy&Co
st&($
millions)&
Total&Subsidy&Cost&across&all&countries&
Cost=Effectiveness)
25&
Cost4Effe
c7vene
ss&($
/Death&Averted
)&Cost4Effec7veness&across&all&countries&(low&elas7city)&
Excludes:&Senegal,&Somalia,&Kenya,&Rwanda,&Zimbabwe,&Sudan,&Ethiopia,&Mauritania,&Namibia,&Swaziland,&Djibou7&
Partial)Subsidy:)No)Subsidy)
26&
Deaths&Averted&over&Five&Years&(low&elas7city)&
Deaths&Averted
&(tho
usands)&
Summary)• Saving&lives&usually&comes&at&a&(diminishing&marginal)&cost:&
• A&universal&subsidy&saves&the&most&lives&and&costs&the&most&• A&par7al&subsidy&or&a&child&targeted&subsidy&with&low&leakage&to&adults&tends&to&be&more&cost4effec7ve&To&maximize&the&number&of&lives&saved&
• To&maximize&the&number&of&lives&saved:&• Expand&access&to&drugs&purchased&at&the&AMFm&nego7ated&price.&&Reduces&risk&of&leakage&across&borders,&lower&nego7ated&prices&and&improve&access&to&QA4ACTs.&
• Tailor&AMFm&to&county&needs:&Spend&$$&on&countries&that&give&the&biggest&bang&for&buck&–&whether&on&child&targeted&subsidy&or&par7al&subsidy&depending&on&local&context.&
27&