[Early 2017] Charity Science Health CEA

	A
1	Notes Sheet
2	General:
3	Enter values in the "Parameters" and "Values Assignment" tabs. After filling in all your input values, view the "Results" tab.
4	Resizing the formula bar in Google Sheets may make it easier to read the contents of explanatory cells
5
6	Color code on "Parameters" and "Values" tabs:
7	Orange indicates parameters that are both uncertain and likely to have major affects on results
8	Pink marks parameters that are more certain or less influential
9
10	Color code on intervention tabs:
11	Blue marks calculations
12	Orange marks values pulled from "Parameters" or "Value Assignments" tabs
13	White marks defult values
14	Pink marks items changed from their default values
15
16	Notes on CSH Model
17
18	We assume that CSH does not send SMS messages for BCG (recommended at birth).
19	Costs are "Based on 2 messages sent for each UIP vaccination time for children under 5. 6 weeks, 10 weeks, 14 weeks, 9 months, 16-24 months, 24-60 months, 60 months." We model an effect on immunization visits at at 6, 10, 14 weeks and at 9 months, as we assume that the effect of booster vaccinations after 9 months are small compared to the effect of the first course of vaccinations. See "Immunization Schedule" tab.
20	We model some, but not all, diseases. See "Immunization Schedule" tab.
21
22	We assume that SMS reminders are sent to every enrollee, regardless of the survival status of the child (i.e. CSH does not receive feedback on whether recipients are still alive). This is unlike the New Incentives program, which pays out incentives ony to enrollees whose children are still alive (to come collect incentives).
23
24	We model the survival outcomes of a cohort of children, including death from all-cause mortality. Hence, in this model, an "under-5 death averted" is a child who survives to 5yo who otherwise would not have survived to 5yo. This is different than our models of antimalarial interventions, in which an "under-5 death averted" represents a child under 5 who does not die of malaria, and otherwise would have died of malaria. We expect the difference between these approaches to be small.
25
26	As India has relatively high estimated rates of vaccination, the effect of the CSH program may be to increase the timeliness of vaccination, rather than the vaccination rate. We have not explicitly modeled the benefits of increasing timeliness of vaccination, and have not seen data on age at vaccination.
27	Vaccination in India differs from Nigeria in ways which affect this CEA (compared to New Incentives): In countries with high rates of vaccination, we assume that disease transmission is lower than in countries with low rates of vaccination (10% yearly chance of infection rather than 30%). India has a nonnegligible rate of MCV2 vaccination, a second dose of measles at 15mo.
28	We do not include costs or benefits from the governmental or hospital persepctive of providing additional vaccines. We expect that while infrastructure costs are for the most part fixed, causing additional vaccine uptake does cause governments to incur small marginal costs. However, we also expect that from the governmental perspective, public health cost savings outweigh the small costs of vaccination. We do not include either these costs or benefits.
29
30	Helpful GiveWell references:
31	Top Charities
32	Other cost-effectiveness analyses (including past years' versions)
33	Discussion of cost-effectiveness as a criteria
34	Blog posts discussing cost-effectiveness
35	Cost-Effectiveness Analysis FAQ
36	Cost-Effectiveness Analysis Video Walkthrough — Make sure to watch the video in HD in full screen to see details
37
38
39
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