Does the estimate of 154 million lives saved by vaccines since 1974 stack up?

In May 2024, The Lancet published an estimate of 154 million lives saved by the Expanded Program on Immunization (EPI) that was launched in 1974. In this piece I show that this estimate doesn’t stack up.

1. Measles vaccines could not have saved 93.7 million lives

Let me begin by looking at the paper’s claim that 93.7 million out of the total of 154 million lives saved were by the measles vaccine. 

According to Our World In Data (OWID), 810,390 people died globally from measles in 1980. Let’s assume that 0.88 million died from measles in 1974. If all measles deaths disappeared from the planet the moment EPI started, the total measles deaths averted between 1974 and 2024 would have been 50 x 0.88 = 44 million. Of course, we must account for global population growth from 4 billion in 1974 to 8.2 billion in 2024. So, the 0.88 million estimate for 2024 should become 1.8 million, and we should add the area of the triangle with height (1.8 – 0.88) million, or 0.92 million, and base 50, which is 23 million. Thus, a total of 67 million measles deaths could possibly have been averted during EPI had measles disappeared. 

The Lancet paper correctly notes that not all these lives would be attributable to vaccines. It attributes 40% to vaccines. So a maximum of 26.8 million deaths (clearly far less than 93.7 million) could have been averted by the measles vaccine – had measles disappeared.

But measles did not disappear. OWID data show that only around 31 million measles deaths were averted during the EPI period. Using Lancet’s rule, 12.4 million lives (40% of 31 million) were thus potentially saved by the measles vaccine – far fewer than the claimed 93.7 million. But I’ll now show that this figure should be even lower.

2. The 40% ratio doesn’t apply to the West

Measles deaths had almost disappeared in the West before the vaccines came in. Only around 500 persons per year were dying of measles in the USA at that time and around 100 each in the UK, France, Denmark and Sweden. But the Lancet paper claims that the measles vaccine saved 15 million lives in the Americas and 6.5 million in Europe. The word, “impossible” comes to mind.

Infant mortality rates (IMR) started declining in the UK from around 1700, driven by increased incomes and falling food prices. Towards 1890, sanitation and clean water came into the mix; and hygiene and clean milk from the early 20th century. Consequently, IMR had collapsed in the UK by more than 90% from 1700 levels by 1940, with an ongoing downward trend. This was before antibiotics (1944) and widespread vaccines (1950s).

Thomas McKeown was one the most prominent analysts of life expectancy in the West. John Bunker summarised McKeown’s findings thus, in a 2001 paper: “Life expectancy had increased by 23 years during the first half [of the 20th] century, but McKeown was able to attribute no more than a year or two to advances in medical care”. Note that “medical care” includes antibiotics, which had a strong impact on bacterial infections, leaving us with less than 2-5% of the decline of IMR in the West from vaccines. The 40% ratio is too high.

3. The 40% ratio can’t apply even to the Third World

The Third World was able to copy Western advances in sanitation, clean water, hygiene and antibiotics long before vaccines came in (e.g. India produced vast quantities of antibiotics from 1956). That’s why India’s IMR declined from around 220 per 1000 births in 1915 to around 130 per 1000 births by 1974, i.e., before EPI. 

Had vaccines contributed 40% of the marginal reduction in IMR in the Third World, the global rate of decline of IMR would have sharply steepened after 1976 (adding a two-year lag) but there was no change to the trend. Even if IMR decline had steepened, that wouldn’t have proved that vaccines caused it. We’d still need to control for coterminous improvements. Surplus milk powder from the USA and Europe was part of foreign aid from the 1960s. And global milk production rose steeply. For example, India’s milk output increased from 17 million tonnes in 1951 to 240 million tonnes today, with a four-fold increase in per capita availability. High yielding crops boosted food supply. And total fertility rates declined, enabling mothers to care better for fewer children. 

If you recall my write-up on the role of clean milk in measles mortality decline in the West, you’ll recognise in the above list many of the same factors which played a pivotal role in the pre-1945 (pre-vaccine) decline of measles mortality in the UK. Moreover, the juxtaposition of child mortality charts from Africa and the UK shows that the rate of mortality decline (-0.35) in the UK between 1890 and 1945 was identical to Africa’s between 1968 to 2023 (-0.34). This corroborates the above evidence, that child mortality decline in Africa from 1968 (and, indeed, from earlier) was driven by the same causes which operated in pre-vaccine UK. 

4. Case of the measles vaccine

The data on measles mortality also force us to reject the idea of vaccines contributing 40% to the observed decline in mortality. By 1988, 70% of the world’s children were vaccinated against measles, so measles deaths should have reduced sharply by 1990. But they didn’t budge – from the previous trend. Public health is always keen to show us charts of the steep decline in measles incidence in the West after the vaccine was introduced, but never any chart of the decline in mortality after vaccines. Because there is none.

After 1988, the global take-up of the measles vaccine stagnated, rising slightly to 80% by 2008. At that point we should have expected measles deaths to flatline. Instead, we see a sharp steepening of the decline in measles mortality from around 1997. Why? By the late 1990s, extreme poverty plummeted globally, enabling more parents to afford milk or alternatives with Vitamin A, driving measles deaths down. The fact that measles deaths dropped like a rock after poverty was eliminated – even as the vaccine rate remained stagnant – shows even more clearly that capitalism, not vaccines, drove the decline. 

In 2023, 83% of children globally received their first dose of the measles vaccine, 17% (340 million) did not. Yet, measles deaths have fallen to just 50,000 per year, a 97% reduction from 1.8 million (recall: the 1974 measles deaths figure extrapolated to 2024 after controlling for population growth). Something other than vaccines has been driving down measles deaths. I believe therefore that measles vaccines might have saved at most, say, 5% of the 31 million deaths averted during EPI.

Conclusion

The Lancet paper is not anchored to reality and should be retracted. Its authors should go back to the drawing board. Note that I am not suggesting that vaccines are without value. Not only was the smallpox vaccine essential for the eradication of the disease, most vaccines clearly reduce the incidence of disease while also saving a few lives. Since such lives saved are mostly of children, vaccines likely pass a cost-benefit test.