ABCDEFGHIJKLMNOPQRSTUVWXYZ
1
2
Proportion of mortality attributed to malaria in areas AMF works vs. the contexts of trials in the Cochrane Pryce et al. 2018 meta-analysis:
3
DRC180%
4
Guinea124%
5
Togo101%
6
Uganda153%
7
Nigeria95%
8
9
Data
10
AMF countries (GBD data from 2019)
1-59 mos all-cause mortality in a year per 1,000 children of that age (GBD)
1-59 mos direct malaria mortality rate per 1,000 person-years (GBD)
Direct malaria mortality as a percent of all-cause mortalityDecline in malaria mortality due to ITN coverageDirect malaria mortality as a percent of all malaria-attributable mortalityDirect malaria mortality after stripping out the effect of ITNsAll-cause mortality after stripping out the effect of ITNsDirect malaria mortality as a percent of all-cause mortality after stripping out the effect of ITNs
11
DRC7.73.139%32%67%4.59.945.3%
12
Guinea14.63.725%32%67%5.417.231.4%
13
Togo8.31.620%32%67%2.49.425.5%
14
Uganda7.32.432%32%67%3.48.938.5%
15
Nigeria15.02.818%32%67%4.117.023.9%
16
17
All-cause under-5 mortality (per 1,000 under-5 person-years)Estimate of protective efficacyMortality reduction (per 1,000 under-5 person-years)Expected reduction in malaria-attributable mortality from ITN coverage in RCTsEstimated malaria-attributable mortalityDirect malaria mortality as a percent of all malaria-attributable mortalityImplied direct malaria mortality rateImplied direct malaria mortality as a percent of all-cause mortality
18
Cochrane Pryce et al. 2018 meta-analysis data3817%6.445%1467%9.525.2%
19
20
21
22
Sensitivity checks
23
In the model above, we use the difference in all-cause mortality rates between the contexts of ITN RCTs and today's contexts as a proxy for the differences we expect in ITNs' capacity to avert deaths in different settings.
For this to be a good proxy, malaria mortality needs to be a relatively stable fraction of all-cause mortality between the RCT contexts and the modern-day AMF contexts.
24
These sensitivity checks are meant to be illustrative rather than definitive. Malaria mortality estimates in low-resource settings tend to rely on low quality data. The sensitivity checks involve many uncertain factors. We encourage individuals engaging deeply with our CEA to experiment with different values for inputs to these sensitivity checks.
25
26
Sensitivity check #1
27
Below, we compare the decline in malaria mortality rates since 2000 to the decline in all-cause mortality rates since 2000. Similar declines in both metrics suggest that we are using a reliable proxy.
28
29
Decline in malaria deaths since 2000 (WMR 2015)60%
30
Percent of decline from 2000 to 2010 attributed to ITNs
24%
31
Adjustment for increasing ITN coverage between 2010 and 2015
30%
32
Percent of decline from 2000 to 2015 attributed to ITNs
32%
33
Decline in malaria deaths since 2000 (with ITNs stripped out)
41%
34
35
Under-5 all-cause mortality rate in SSA in 2000 per 1,000 children (according to GBD)
34.18
36
Under-5 all-cause mortality rate in SSA in 2015 per 1,000 children (according to GBD)
17.87
37
How much to strip out of that because of ITNs?25%
38
Under-5 all-cause mortality in SSA in 2015 per 1,000 children (with ITNs stripped out)
21.9475
39
Decline in all-cause mortality rate since 2000 (with ITNs stripped out)
36%
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100