Is vaccination good economics as well as good politics?

Vaccinating children is obviously good politics, demonstrating what can be achieved with aid. There was great celebration when the Global Alliance for Vaccines and Immunisation (GAVI) raised over $US 4 bn at its pledging conference in London, back in June. David Cameron even received a standing ovation for pledging some $US.1.3bn.  250 million children will be vaccinated. 100 million lives will be saved. Fantastic.

The immediate economics is also highly positive. Look at the 2008 summary, below, of health ‘best-buys’ by the Disease Control Priorities Project. Vaccinating children against major childhood diseases comes top of the list. Again, a fantastic investment.

1. Vaccinate children against major childhood diseases, including tuberculosis, diphtheria, whooping cough, tetanus, polio, and measles (the traditional expanded program of immunization).
2. Monitor children’s health to prevent or, if necessary, treat childhood pneumonia, diarrhea, and malaria.
3. Tax tobacco products to increase consumers’ costs by at least one-third and reduce cases of cardiovascular disease, cancer, and respiratory disease (see Box 1 on page 4).
4. Prevent the spread of HIV through a coordinated approach that includes: promoting 100 percent condom use among populations at high risk of infection; treating other sexually transmitted infections; providing antiretroviral medications to pregnant women; and offering voluntary HIV counseling and testing.
5. Give   children  and pregnant  women  essental nutrients, including vitamin A, iron, and iodine, to prevent maternal anemia, infant deaths, and longterm health problems.
6. Provide insecticide-treated bed nets, household spraying of insecticides, and preventive treatment for pregnant women to drastically reduce malaria in areas where it is endemic.
7. Enforce traffic regulations and install speed bumps at dangerous intersections to reduce traffic-related injuries.
8. Treat tuberculosis  patients  with   short- course chemotherapy to cure infected people and prevent new infections (see Box 2 on page 5).
9. Teach mothers and train birth attendants to keep newborns warm and clean to reduce illness and death.
10. Promote the use of aspirin and other inexpensive drugs to prevent and treat heart attack and stroke.

Hang on a minute, though. The major pledges to GAVI were not for the traditional ‘expanded programme of immunization’ (EPI), but mainly for new vaccines, especially against rotavirus and pneumococcal infection. The press release and other coverage confirms that this is the case. The figure below shows that new vaccines are the biggest growth areas. Are the economics then as clear?

What led me to ask this question was the juxtaposition of the figure raised at the London conference, $US 4.3 bn, with the number of children to be vaccinated, 250m, and the number of lives to be saved, 4 m. Obviously, it is wonderful to save 4 million young lives, but an immediate mental calculation, dividing $4.3 bn by 4 m, told me that each life saved was going to cost over $US 1000. That seemed a lot, especially given what we know about health budgets in developing countries (less than $US 100 per capita for the poorest 100 countries in the world, even in purchasing power parity dollars) and about the extremely low cost of some other interventions – for example, micronutrient supplementation. Every investment has an opportunity cost. Could even more lives have been saved by raising the same amount of money for nutrition?

The GAVI event also triggered a memory – of being invited to a lunch in London a few years ago, when I was Director of ODI, the purpose of which was for a pharmaceutical company to discuss with development people the following issue: ‘at great cost to ourselves, we have invented some new vaccines, including against rotavirus, and we want to persuade donors to pay for them; they are quite expensive; how do we do that?’.

I remember making the argument that we development specialists believed very strongly in rational public expenditure allocation, that we helped to design medium term expenditure frameworks, that we opposed vertical programmes that parachuted money into health ministries with no regard for country ownership, national priorities or the need to build health systems; and that therefore the logical thing for the pharmaceutical companies to do was to make a careful argument for cost-effectiveness, based on sector programmes at health ministry level, within overall national public expenditure ceilings.

You can see that I was reflecting faithfully, and with conviction, I might say, the conventional wisdom of ODI’s public expenditure specialists.  I also had in mind the famous case of the Uganda health ministry back in 2002, being offered more money from the Global Fund for HIV/AIDSHuman Immuno-Deficiency Virus/Aquired Immuneo-Deficiency Syndrome than the whole health budget of the country, and the horrified reaction of the Ministry of Finance at the distortion this implied to overall spending patterns.

My thinking then was consistent with scepticism about what later I have come to call Results 1.0, the view that maintaining political support for aid requires such a strong focus on results that ‘simple’ interventions, like vaccination, trump more complex and equally important ones like support to the institutions required for better governance. As readers of a previous contribution will know, I think Andrew Mitchell is managing this conundrum well, with a subtle agenda I have called Results 2.0.

The interesting part of this story, though, is the reaction of the pharmaceutical company representatives, senior people all, who looked at me as though I was both deranged and naïve beyond belief. That, their expressions said, is not how it works.

You can understand then, my interest being awoken, on finding, only a few years later, that the vertical fund created to fund vaccination was making a major commitment to new vaccines, at what seemed a high cost per life saved. Was this the triumph of politics over economics? Or had the economics changed?

I raised these questions at an ODI/IDS meeting on results, and again at a World Bank seminar in London in September. Some interesting points have come up.

First, of course, the cost per life saved is a crude way to think about the economics, and it is more accurate to use Disability-Adjusted Life Years (DALYS), appropriately valued and discounted.

Second, a DALY perspective confirms that traditional EPI vaccination is highly cost-effective, at less than $US 10 per DALY averted.

Third, the new vaccines are indeed much more expensive, though coming down in price, which makes a literature review and comparative analysis quite complicated. The GAVI page on cost-effectiveness makes statements about cost-effectiveness, but does not give numbers. See: http://www.gavialliance.org/about/value/cost-effective/. The Disease Control Priority Project cites rotavirus is cited as a “high-cost intervention” in both South Asia ($US 500-7000 per DALY averted) and SSASub-Saharan Africa ( $US 500-1700). There are no data for pneumococcal vaccinations, as far as I can see. However, there are plenty of refs on the web to DALYs and pneumococcal vaccine. The Lancet paper cited by GAVI says $US 100 per DALY averted. A 2011 paper comparing different models has figures from $US 767 - $US 4000 (see http://www.biomedcentral.com/1741-7015/9/53).

Fourth, micronutrient supplementation is not the only nutrition intervention to score highly in comparative tests of cost-effectiveness. Work for the Copenhagen Consensus shows many nutrition interventions with cost per DALY saved below $US 20, or in the range $US 20-100. See http://thousanddays.org/wp-content/uploads/2011/05/Copenhagen-Consensus-2008-summary.pdf.

Fifth, there are many other cost-effective investments available to save lives and reduce poverty – including insecticide-impregnated bednets, measures to provide clean water and better sanitation, improved stoves to reduce smoke inhalation, and many others. I brought some lists together when writing in 2008 about the need to move from risk frameworks to opportunities frameworks.

However . . .

Sixth, it is not surprising that new vaccines are more expensive than well-established ones, since R and D costs have to be amortised. Eventually, the cost of new vaccines will come down and be competitive. Owen Barder, for example, has argued that the extra initial price is worth paying to cover the costs of development (and investment in production) for new vaccines, provided that in the long run prices fall down to the social marginal cost. That’s interesting and sounds like a standard cost-benefit problem. We’re comparing two investments, one of which (new vaccines) is expensive now but will be cheaper later, the other of which is relatively cheap now and in the future. Which we choose depends on the relative prices now and later, and on the discount rate. Without carrying out detailed calculations, Figure 1 below illustrates the issue. It shows the actual costs over a 50-year period, with a nutrition intervention constant in price at $US 20 per DALY, and two options for vaccination, the first at $US 100, declining by 10% a year, the second with the same initial cost, declining at 5% a year. The vaccinations are eventually cheaper. However, if a discount rate of 10% is applied, the nutrition intervention still comes out top, because the initial cost is so much lower. Remember, these numbers are hyothetical.

Figure 1

What do I conclude from all this? Well, the politicians may think the politics is complicated, but, speaking as an economist, it seems to me that the economics is pretty complicated too. An awful lot depends on which method is used to calculate cost-effectiveness, and on the specific valuations and discount rates applied. There is no excuse, however, for advisers not to carry out the analysis and present ministers with costed options.

What should ministers then do? Logically, they should start at the top and work down. Practically, they will want to think about the politics as well as the economics, and about feasibility and adminsitrative simplicity. Probably they will find good reasons to invest in vaccines, both old and `` They will also find many good reasons to invest in other interventions to save the lives of poor children. Nutrition will be high on the list.