ABCDEFGHIJKLMNOPQRSTUVWXYZ
1
Best guess for endline results
2
Measles
Other vaccinations
3
Pre-midline prior for percentage point increase in vaccine uptake13.00%15.00%
4
Best guess of midline results (pp increase in vaccine uptake)7.97%12.58%
5
Weight on prior25.00%25.00%
6
Post-midline best guess of endline RCT result (pp increase in vaccine uptake)9.23%13.19%
7
8
Estimated effects based on midline results
9
10
Option 1: Medium estimate using baseline & midline data
11
MeaslesBCGAny PentaFull Penta
12
Baseline survey: coverage rate in Katsina and Zamfara16.00%24.70%21.40%5.70%
13
Midline: baseline mean control volume26.4720.3929.3726.44
14
Midline: baseline mean treatment/control difference1.035.942.721.76
15
Midline: mean change in control clinic volume, baseline to midline -62.67-0.46-0.45
16
Midline: mean additional change in vaccination volume in treatment clinics28.5936.7733.0832.32
17
Midline: counterfactual vaccination volume in treatment clinics21.52931.6327.75
18
Estimated excess volume of out-of-catchment vaccinations in treatment clinics8.518.518.518.51
19
Key assumption: what % of excess volume is driven by improved recordkeeping? 40.00%40.00%40.00%40.00%
20
Mean change in vaccination volume, recordkeeping effect stripped out17.15422.06219.84819.392
21
Excess volume of out-of-catchment vaccinations, recordkeeping effect stripped out5.1065.1065.1065.106
22
Estimated change in counterfactual coverage rates since baseline-22.23%11.43%-1.50%-1.65%
23
Estimated counterfactual coverage rate12%28%21%6%
24
Mean change in within-catchment vaccination volume, recordkeeping effect stripped out12.04816.95614.74214.286
25
Implied percentage point change in coverage rate6.97%16.09%9.82%2.89%
26
Average of BCG and Penta 1 (to serve as non-measles input for NI CEA)12.96%
27
Average of measles, BCG, and Penta 110.96%
28
29
Medium Option 2: same as option 1, except uses baseline coverage rates instead of counterfactual midline rates
30
Implied percentage point change in coverage rate8.97%14.44%9.97%2.93%
*Measles is sensitive to this choice because control group volumes fell between baseline and midline
31
Average of BCG and Penta 1 (to serve as non-measles input for NI CEA)12.21%
32
Average of measles, BCG, and Penta 111.13%
33
34
3. High-side midline estimate that assumes no recordkeeping bias (but strips out out-of-catchment vaccinations)
35
Key assumption: what % of excess volume is driven by improved recordkeeping? 0.00%0.00%0.00%0.00%
36
Mean change in within-catchment vaccination volume (OOC vaccinations stripped out)20.0828.2624.5723.81
37
Implied percentage point change in coverage rate13.28%25.45%16.50%4.85%
*Uses average of baseline and implied counterfactual coverage
38
Average of BCG and Penta 1 (to serve as non-measles input for NI CEA)20.97%
39
40
4. Low-side midline estimate that assumes a large recordkeeping bias
41
Key assumption: what % of excess volume is driven by improved recordkeeping? 80.00%80.00%80.00%80.00%
42
Mean change in vaccination volume, recordkeeping effect stripped out5.7187.3546.6166.464
43
Excess volume of out-of-catchment vaccinations, recordkeeping effect stripped out1.7021.7021.7021.702
44
Implied percentage point change in coverage rate2.7%5.1%3.3%1.0%
*Uses average of baseline and implied counterfactual coverage
45
Average of BCG and Penta 1 (to serve as non-measles input for NI CEA)4.2%
46
47
5. Upper bound estimates of effect sizes based on baseline "switchers"
48
% of those skipping vaccination in baseline survey who we believe could be incentivized to switch64.00%64.00%64.00%64.00%
49
% that skipped vaccination at baseline84.00%75.30%78.60%94.30%
50
Upper bound percentage point increase in immunization (% of switchers among non-vaccinators)54%48%50%60%
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