Vitamin A: how a two-cent capsule became a $3,500 life

Vitamin A deficiency weakens a child’s immune system. In places where diets lack it, children die more often from measles, diarrhea and other infections they would otherwise survive. The fix is a high-dose capsule, twice a year, from six months to five years old. The capsule costs a few cents.

Helen Keller Intl supports governments in sub-Saharan Africa to run those campaigns. GiveWell estimates the cost at about $3,500 per life saved, its lowest figure among its top charities. Here is how a two-cent capsule becomes a $3,500 life.

Step 1: the capsule is the cheap part

Delivering it is not. A campaign has to reach every eligible child in a district, twice a year, through health workers who need training, transport, record-keeping and supervision. GiveWell’s estimate of the full delivered cost is roughly $1 to $2 per child per round, including the government’s share, not just the charity’s.

Step 2: most children would have lived anyway

Under-5 mortality in the countries where these campaigns run is high by global standards, perhaps 5 to 10 percent of children, but that still means 90-plus percent survive without the capsule. The supplement lowers the death rate among those who get it; it does not save each one.

Step 3: how much it lowers the death rate is contested

The Cochrane meta-analysis (2017, 47 trials, about 1.2 million children) found vitamin A supplementation reduced all-cause mortality by about 12% in children aged six months to five years.

Against that stands DEVTA, a trial of about one million children in northern India published in 2013, which found a much smaller effect, around 4%, that was not statistically significant. DEVTA is the single largest trial ever run on the question, and its null-ish result has never been fully explained. Possibilities include lower deficiency rates in the study area, lower coverage than reported, or simple chance.

GiveWell’s model uses an effect size well below the Cochrane average to reflect DEVTA, and then adjusts further for how prevalent deficiency is in each specific country today, since the benefit is larger where deficiency is worse.

Step 4: multiply

Take a district where 1 to 2 percent of children in the age range would die in a given year from causes vitamin A affects, apply a mortality reduction of 5 to 12 percent depending on your read of the evidence, and you save one life for every few thousand children covered. At $1 to $2 per child per round, two rounds a year, the cost per life saved lands in the low thousands. GiveWell’s published figure, averaged over its 2022 to 2024 grants and rounded, is $3,500.

Why it still ranks first

The mechanism is well understood and the deficiency is measurable. Programs go where deficiency is documented.
Governments already run the campaigns. The charity’s money buys coverage and quality, not a parallel system.
The downside is small. If DEVTA is right and the effect is 4%, the cost per life roughly triples to something near $10,000, still in the range of good programs.

What would change the picture

Fortification of cooking oil and flour with vitamin A is spreading; where it succeeds, capsule campaigns become redundant. A donor should expect this program to shrink over a decade, which is the right kind of ending for a health intervention.

Method notes

Cost per life saved is GiveWell’s 2022 to 2024 average. Per-child delivery costs and mortality rates are rounded from GiveWell’s model inputs. The worked arithmetic in step 4 is ours and is meant to show the order of magnitude, not reproduce GiveWell’s spreadsheet.

Sources

Figures come from the cited sources as of the dates shown. Cost-effectiveness estimates change as programs and evidence change; check the source before giving. We are not financial advisers and receive nothing from any charity named here.