Medical Research - Part 3

You do not have Javascript enabled. Some elements of this website may not work correctly.

Notice about research

Giving What We Can no longer conducts our own research into charities and cause areas. Instead, we're relying on the work of organisations including J-PAL, GiveWell, and the Open Philanthropy Project, which are in a better position to provide more comprehensive research coverage.

These research reports represent our thinking as of late 2016, and much of the information will be relevant for making decisions about how to donate as effectively as possible. However we are not updating them and the information may therefore be out of date.

  • Published 21 Sep 2015
  • Updated 3 Mar 2020

0.0.1. See also Part 1 and Part 2.

1. Appendix 1: Disease-specific evidence of need

This section notes disease specific evidence that I turned up in my literature review. I did not try to do a thorough literature review into these questions, but I thought it worth recording what I did find.

1.1. Some words of caution about assessing disease-specific need

One thing to note is that we are not interested solely in the current disease burden. Disease burdens can change. For instance, the pre-2013 disease burden of Ebola was very low, but an epidemic was a known risk. During this time, there were vaccines in late stages of development, but they were not brought to market because there was no current disease burden. This indicates that current disease burdens are not necessarily a good measure of potential disease burdens. We should therefore be proactive, and consider potential future disease burdens.[1]

1.2. General resources for assessing disease-specific need.

One very simplistic analysis of greatest need is to compare the amount of research currently being carried out (proxied by the number of citations there are on a particular disease) to the disease burden.[2] The Global Forum for Health research, accounting mainly for this consideration, argued that more investment was needed in ‘acute respiratory infections, diarrhoeal diseases, cardiovascular diseases, mental health, tuberculosis, tropical disease, perinatal conditions and HIV/AIDS’. However, this analysis is not particularly helpful, because, though it notes that all of the diseases that we are studying are relatively underfunded, it does not differentiate between them.[3]

Moreover, such an analysis crucially omits the evidence on scientific promise, and current control methods, the very considerations that this section is focusing on. There are some general resources on the current control methods available for a disease. Plotkin, Mahmoud, & Farrar, (2015) classify different diseases according to whether there is a current effective vaccine for them.[4]

Turning to the scientific challenges, the Global IDEA Scientific advisory committee details the needs within each disease area.[5] Meanwhile, a BVGH report provides detail on the current state of vaccine development for a range of 17 relevant diseases.[6]

1.3. Disease-specific resources

The structure here varies according to the research that I encountered on each topic.

1.3.1. Malaria

There are a range of possible technologies that would be useful in combatting malaria, including diagnostic tools, vaccines, drugs, and insecticides. Malaria is a particularly difficult disease to develop technologies for, because it is a parasite, and so more biologically complex. This means that it has more redundant genes, and more mutations, and it is difficult to develop vaccines or drugs that affect a sufficient amount of the organism’s functions, or to develop accurate diagnostics, due to variation in the parasite.[7] According to G-FINDER, the total spent worldwide on research into malaria was $549m.[8] Drug Development

Although there are drugs for malaria, they are only partly effective, and this effectiveness is declining due to increasing resistance, particularly in South-East Asia, where drug use is highest. Monotherapies (with one active ingredient) are cheaper, but less effective than, combination therapies. However, combination therapy resistance is also growing, and new drugs are needed to replace the current combination therapies.[9]

Drug development proceeds from a random selection of compounds: a large set of compounds is tested for efficacy in the laboratory, and then in human subjects, with fewer compounds reaching each subsequent stage. Therefore, if the rate at which new compounds can be screened increases, the time to vaccination would decrease roughly linearly. Manary estimates that around 1.25m compounds need to be screened for each drug that can be brought to market. Since 250,000-500,000 compounds are screened annually, at this rate it should take between 3-5 years to develop a new antimalarial drug. However, since there are not enough new libraries of compounds to find potential drugs, fewer compounds are being tested, and it may take between 12-15 years to find a new drug. Manary also gives a breakdown of costs needed to start a new drug-development laboratory, which come in at $5m up front, followed by around $2.5m/year.[10]

Despite the clear importance and relative tractability of this area, however, Towse notes that there is already high investment in drug development, largely because there are significant developed-world markets for the drugs (for travellers and the military).[11] The area is therefore not particularly neglected. Vaccine Development

There are currently no vaccines being widely used against malaria. However, one candidate, Mosquirix, has passed through all clinical trials, and received approval from the European Medicines Agency. The vaccine appears to be between 27% and 45% efficacious over 18 months. Immunity drops somewhat after that, and immunity is lower in younger children. It was expected that the WHO would recommend Mosquirix in the future, and that after this the roll-out of the vaccine would begin[12] - likely organised by the Global Alliance for Vaccine and Immunisation (GAVI).[13] In October 2015, the WHO’s Strategic Advisory Group of Experts recommended that pilot programs begin to determine how to most effectively use the vaccine, but it remains to be seen whether this will lead to a much larger roll-out.[14]

However, this does not signal the end of malaria vaccine development: due to the low efficacy of the vaccine, the search for a better vaccine is still worth pursuing. Developing vaccines for Malaria is particularly difficult because Malaria is a parasite, as discussed above.[15] Vector Controls

The current main vector control is the long-lasting insecticide-treated bed net (LLIN), which is a highly cost-effective means of preventing the transmission of malaria. In the past DDT, an insecticide, was used widely, and was effective at eradicating malaria in many regions, though not in tropical regions. DDT use has since been reduced due to environmental concerns.

The WHO argues that new insecticides are needed to ensure the continued efficacy of LLINs, and indoor insecticide spraying.[16] New insecticide developments are likely to come from agricultural rather than medical research.[17] Diagnostics

Malaria can be diagnosed using microscopy, which is simple and effective, but requires facilities, and trained personnel. Therefore, much diagnosis is based on clinical signs and symptoms, or rapid diagnostic tests, which are less reliable. This has led to over-prescription of malarial drugs. Priorities

The Global Health Technologies Coalition argued in 2011 that diagnostics and vector controls are most in need of new funding. Vaccines are currently well funded, as they are in the cheaper early stages of trials. However, as they enter stage 3 trials, which are expensive because they involve large numbers of human subjects, they will need more funding. Overall, they argue that a steady increase in funding is needed for a number of years, before projects begin to reach their targets, when can begin to be cut.[18] The Medicines for Malaria Venture also estimates that funding for diagnostics and vector controls needs to be doubled, whereas drugs and vaccines need only steady funding increases.[19] In contrast, Micah Manary, speaking to GiveWell, notes that there are currently available drugs and vector controls, but that we lack an effective vaccine, thus implying that vaccine research is a higher priority.[20] Many of these resources also contain useful discussions of the specifics of developing different types of technology. In conclusion, it remains unclear what the priorities for malaria spending should be.

1.3.2. HIV/AIDS

HIV/AIDS is another disease that presents particular scientific problems for drug and vaccine development.[21] Diagnostics

There exist inexpensive, accurate diagnostics which test whether someone has HIV. These tests don’t always work, but if they fail there are good alternatives. There are also tests, which take a longer time, to determine the extent of the disease and optimum treatment options.[22] Vaccines

There are currently no vaccines effective against HIV. However, there are known methods for reducing infection rates, for instance, male circumcision (which may reduce risk by up to 60%). In addition, anti-retroviral treatments reduce viral load, and so reduce infection rates. These methods have been largely successful in controlling the HIV epidemic. One problem with a vaccine is that it may lead to more risky behaviour, which might reduce the reduction in transmission rates.

However, a vaccine for HIV remains promising, as the Copenhagen Consensus Centre’s cost-benefit analysis (discussed above), indicates.[23] Drugs

Antiretrovirals slow the progress of the disease, but more effective drugs would be useful. [24]

1.3.3. Tuberculosis Diagnostics

The few diagnostic techniques that exist are expensive. Cheaper, more rapid tests are needed, in part in order to prevent the spread of drug-resistant strains of TB.[25] Drugs

There are a wide variety of drugs available for treating TB, however drug-resistance is a continued risk. This may be somewhat countered by early diagnosis and ensuring full treatment, but new drugs will also be useful. Vaccines

The BCG vaccine against TB is widely used, but is only partially effective, and has therefore failed to control the TB epidemic.[26] There is therefore room for a more efficacious vaccine to replace BCG, or a booster vaccine, or a combination of both: these are estimated to be highly cost-effective, though there are caveats, as discussed above.[27] Though there are potential vaccine candidates, there has been a lack of satisfactory pre-requisite technologies that help vaccine development.[28]

1.3.4. Neglected Tropical Diseases

This set of diseases encompasses lymphatic filariasis, onchocerciasis, schistosomiasis, soil transmitted helminths, human African Trypanosomiasis, Chagas disease, and Dengue and others. Since there are drugs that are effective against several of the diseases, one of the main focuses is on dosages, epidemiology, and co-infections. However, there is also a need for developing new drugs.[29] Dengue

There is currently no drug or vaccine for Dengue, and therefore vector control is the key intervention. However, continued investment in vector controls is needed.[30] Lymphatic filariasis

There are already drugs that treat filariasis, but there is a need for more social science research to find the optimal uses for existing diagnostics and drugs.[31] In addition, some drugs with additional features are being developed.[32] Visceral Leishmaniasis

New drugs and diagnostics have made eradication of visceral leishmaniasis a possibility. The main needs for further research are in assessing the most effective eradication methods.[33] Acute Respiratory infections

Vaccines exist for pneumococcal infections: indeed, a new vaccine, more effective in developing countries, was launched in 2010.[34] However, antibiotic resistance, and the costs of diagnosis mean that investment remains worthwhile.[35] Schistosomiasis

Praziquantel, the leading drug against Schistosomiasis, is not completely effective at all stages of its lifecycle. This may be a reason for pursuing further research.[36]

1.3.5. Which of these is most promising?

I have not listed all of the diseases that may be relevant, and my analysis has been (deliberately) cursory. It is not possible to give an assessment of which of these diseases are more promising without a deeper, more scientifically informed, analysis, such as what OpenPhil is carrying out.[37]

2. Appendix 2: How should one fund research into neglected tropical diseases?

This section is intended to be an introduction to some of the considerations and existing resources on the most effective funding methods. It is unclear which of these methods is most promising.

2.1. A concern about marginal impact

GiveWell note that most funders of medical research more generally have large budgets, and claim that ‘It’s reasonable to ask how much value a new funder – even a relatively large one – can add in this context’.[38] Whilst the field of tropical disease research is, as I argued above, more neglected, there are still a number of large foundations, and funding for several diseases is on the scale of hundreds of millions of dollars.[39] Additionally, funding the development of a new drug may cost close to a billion dollars (see Appendix  4).

For these reasons, it is difficult to imagine a marginal dollar having any impact. However, as Macaskill argues at several points in Doing Good Better, this appears to only increase the riskiness of the donation, rather than reducing its expected impact. Most of the time, the additional dollar will not make any difference, but sometimes (very, very, rarely) it will allow a product development partnership to fund a new research programme - a massive impact. The chance of this huge effect means that the expected value of a marginal donation may well still be high. Very roughly, we might think of the expected impact of a marginal dollar being equal to the average impact per dollar of a marginal research programme. For instance, assume the marginal research programme costs $800m, and averts 80,000,000 DALYs, If we assume that each marginal dollar has an equal chance of being the one that causes the programme to happen, then in expectation, the expected impact of the dollar is equal to $80,000,000/800,000,000 DALYs=$10/DALY, which is the average impact of the marginal research programme. Therefore, if the marginal research programme is effective, the impact of a marginal dollar will remain high, in expectation, even though it is likely that the dollar will have no impact.

Even if this is true, this may be a cost if we want to be sure that our money is having some impact, and we’re unwilling to gamble on large, low-probability impacts. The risk is particularly great for research because not only the impact of funding, but the productivity of research, is risky. We are unsure if our marginal dollar will cause a new program to be funded, but we’re also unsure about whether that new program will actually find anything useful, since research is a risky business.

With enough knowledge, however, this could work to our advantage. With sufficient scientific knowledge, and a good understanding of the funding system, it may be possible to target dollars where they will make programmes just possible. If the $800m programme is $10,000 away being funded, we could have a massive impact by providing that $10,000. Moreover, if we were sure that the funding gap was $10,000, our impact could be quite reliably high. However, it may be very difficult to get enough knowledge for this strategy to work.

2.2. Assessing different funding types

There are two broad types of funding. Push funding is simpler: it involves paying for laboratories and scientists to pursue a particular piece of research. Pull funding instead involves providing (generally financial) incentives for research groups to produce a certain product. Another way of looking at it is to see that both mechanisms are trying to increase the amount of research done: push mechanisms attempt to pull the supply curve down, by funding some research costs, whilst pull mechanisms attempt to push the demand curve up, by creating more incentives. Theoretically, both approaches should increase the supply of research.

I will now delve a little deeper into the various alternatives, before evaluating how they compare.

2.2.1. The effectiveness of push mechanisms

Push mechanisms fund research programmes. In the public sector this involves creating new programmes, whilst in the private sector it simply alleviates some of the costs of developing a new drug, making a research program viable.

One general problem with push mechanisms is that they do not provide researchers with an incentive to be cost-efficient: to reduce costs, and focus on the most promising research paths. Of course, researchers, and even firms, also have altruistic motivations: they want to find a new drug in order to help people. But they also have other motivations: researchers may prefer to work on academically interesting options, rather than more promising but less intellectually satisfying approaches; firms may wish to slow the development of a vaccine for a particular disease, so that their drugs can make more money.  Both groups may have incentives to increase costs, so as to gain more funding.[40] This failure to incentivise cost-effective work may therefore reduce the cost-effectiveness of the research compared to pull mechanisms.

However, there are reasons for thinking that push programs can face lower costs compared to commercial production (as discussed in appendix 4), since nonprofits can negotiate cheaper rates for equipment, and other inputs.[41] Fund new studies

One type of funding is to fund research directly (for instance, to fund research in a university). The returns to this sort of funding are likely to be highly variable. Some of the factors that cause impact to vary are whether such research substitutes or complements for private sector research, and whether public sector researchers are connected to private sector researchers, with complementarity and connectedness seeming to cause a greater impact.[42]

A potential problem with this option is that, since particular studies are funded, the choice of studies becomes very important. The prioritisation process may be too conservative, preferring conventional wisdom over creativity, and thereby stifling potentially promising research methods. Additionally, it may be that the outlook is too short term, and studies are unrealistically expected to quickly reach results, stifling promising but more complicated approaches. Finally, it may be that grants are given to very specific project. This might prevent investigators switching to a more promising research avenue, if they find that they cannot make progress on their grant project. These concerns are worrying. However they may be justified by the need to closely specify that grant money must be spent on the most promising areas. There is a tradeoff between closely specifying the parameters of a study, to solve the incentive problem noted above, and allowing investigators enough space to pursue what they perceive to be the best decisions. The administrators of these schemes are likely to be aware of this tradeoff. Fund research capacity

An alternative push strategy is to fund research capacity: to ensure that projects have the infrastructure and human resources that they need to be successful. This decreases the costs of technologies and labour for research projects, and thereby stimulates them. This has become one of the focuses of health research policy in developed countries, where capacity improvements are particularly needed.[43]

GiveWell have also investigated improving the efficacy and reliability of scientific research as a promising area.[44] PDPs

So far we have discussed the funding of public institutions. Public-private product development partnerships (PDPs) are collaborations between sectors. As Appendix 3 shows, most of the largest and most successful organisations working on low-income disease research are PDPs.

2.2.2. The effectiveness of pull mechanisms

Pull mechanisms all share the common feature that they incentivise for successful product development, and often provide particular incentives for the first team to develop a successful product. Whilst this incentivises researchers to work on the most promising research areas, it is also risky for the firms: there is a chance that they will fail to develop a product, or that another firm will develop it first. This means that there may need to be relatively high levels of incentivisation in order to induce investment from firms.[45]

Another general problem with pull mechanisms is specifying the desired product closely enough that firms create an appropriate product.[46] Fund prizes for successful products

One option is to simply fund cash prizes for developing drugs, an approach used by the Health Impact Fund, which aims to reward companies for producing drugs according to their estimates of the health impact of the drug.[47]

However, it is unclear whether there would be much of a response from companies to such a prize: this route may need to be demonstrated as viable before companies are willing to participate.[48]

By funding prizes only for a sufficiently successful completed product, the risk to firms is also increased, and firms with lower capitalisation, like biotech firms, may be excluded altogether, since they cannot sustain funding for long enough to reach a complete solution. One solution to this is to have a tournament: to reward the company that produces the best drug by a particular date. However, it may be difficult to define the criteria by which to judge the best product. Another option is to provide small regular prizes for reaching milestones on the development of a product.[49] Pre-commit to buying products

Advanced Market Commitments (AMCs) are legal commitments to buy a certain number of a product, at a certain price, once the product has met a technical specification. There has only been one previous AMC, which successfully brought the pneumococcal vaccine, which was already in a late stage of development, to market. [50]

Like other pull mechanisms, AMCs largely solve the incentive problem, however they are difficult to plan. In particular, it is likely to be difficult to set an appropriate technical specification, or to set an appropriate price in advance of knowing the details (and production costs) of the product. Relatedly, it is difficult to estimate how much money is needed for an AMC to incentivise investment, particularly in early stages of development, when scientific difficulties are not fully understood. If the AMC is set too low, it may have no effect, whereas if it is set too high, additional resources may be simply transferred to the developing company, without improving incentivisation.[51] For this reason, AMCs may be more appropriate to products which are late in the development process, like the pneumococcal vaccine.[52]

Compared to other pull mechanisms, the AMC has two key advantages. First, by organising a purchase commitment, it makes it quicker and easier to distribute a vaccine, potentially reducing the lag between drug approval and distribution. Second, if it includes provisions for later entrants to the market, it can incentivise late entrants, and thereby reduce risk to firms, and encourage more competition, which may drive prices down in the long run.[53]

Stronger, but less plausible claims have been made for AMCs. Proponents have claimed that AMCs are costless until the prize is claimed, but this seems to ignore the cost of capital, or the possibility of diverting attention from alternative funding mechanisms.[54] (Lobbying for) patent extensions / transferable property right awards /transferable fast track

Another possible incentive is to encourage the government to offer firms intellectual property advantages in exchange for producing drugs in this area. Patent extensions and transferable property rights allow a firm to extend their property rights, and thereby increase profits, from a drug of their choice. Transferable fast track review would expedite the review process for a potential drug of the company’s choice, thereby bringing the drug to market more quickly, and increasing firm profits.[55]

Though rather technical, all of these techniques seem to be treated as credible by firms. Moreover, they are free to the government, since they require no extra resources, but merely legal changes.

One cost of patent extensions is that it makes drugs more expensive for longer, thereby possibly excluding some people from the benefits of the drug. It is unclear how significant this disadvantage is.

2.2.3. When should these mechanisms be used?

It appears that all of these options seem useful, to some extent or another. Since push and pull funding tend to complement each other, we need not pick a single funding type exclusively, but can create a portfolio of complementary techniques. Some techniques will be more useful in different situations: in particular, AMCs look promising for later-stage drugs, where they can be better calibrated, and facilitate distribution. At an earlier stage, push mechanisms may be more successful at stimulating investment.[56]

However, I have not done a sufficiently detailed analysis to draw firm conclusions on any of these issues.[57]

3. Appendix 3: What organisations can one support?

Once again, my purpose will not be to assess organisations, but merely to provide a list of organisations that I encountered, to serve as a resource for future research.

Deworm The World, which carries out some research work, is already recommended by Giving What We Can and GiveWell.[58] There are a number of global product development partnerships, generally with a specific disease focus, such as the International AIDS Vaccine Initiative, Malaria Vaccine Initiative, TB Vaccine Initiative, and the Drugs for Neglected Diseases Initiative.[59] PATH, an umbrella organisation, funds research not only into vaccines, but drugs, diagnostics, and technologies.[60] BIO Ventures for Global Health provides training and institutional support for global health drug development.[61] The WHO’s Tropical Disease Research programme works with these and other partners to encourage investment into tropical disease research.[62] TI Pharma enables public-private partnerships.[63] ANDI supports product development in African countries.[64]

Another option is to set up a new organisation.[65]

4. Appendix 4: Evidence on the cost of drug development

The most widely used estimate of the average cost of drug development is $802m ($403m, if one fails to account for the cost of capital). This figure accounts for the capital cost that firms face. It estimates the costs incurred for a large random sample of drugs brought to market, including the costs of related drugs that were not brought to market.[66]

This estimate might be biased by the fact that more scientifically simple drugs are more likely to be invested in first: this suggests that future drugs of the same class may more expensive than past development costs suggest. Further, this figure may exclude the cost of primary research, which may have contributed to finding the drug. These factors mean that the figure may be an underestimate of the true cost for developing new drugs.

Moreover, it is questionable whether this analysis is even relevant for our purposes, since it examines the wrong reference class. First, it examines drug development costs in a private context, whereas most drugs for tropical diseases are developed by public-private product development partnerships (PDPs). PDPs may face lower costs than private development, in part because some of their costs are covered by donations of expertise and equipment.[67] Second, developing drugs for diseases prevalent in low-income countries may be cheaper, on average, than developing drugs for diseases prevalent in high-income countries. Because different types of product are developed in low income countries, fewer clinical trials are needed, and there are lower costs per patient in trial (due to lower personnel/equipment costs in low-income countries, where many of the trials are carried out).

In fact, the costs of developing products for tropical diseases have been estimated to be between $150m and $178m. One reason that this estimate is so low is that it neglects capital costs (since capital costs may be less relevant when considering public spending). However, even accounting only for out-of-pocket costs (that is, disregarding the cost of capital), drug development is estimated to be half of DiMasi et al.’s figure. This is partly because the different types of product developed mean that fewer clinical trials are needed, partly because there are lower costs per patient in trial (due to lower personnel/equipment costs in low-income countries, where many of the trials are carried out), and partly due to in-kind donations reducing costs.[68]

DiMasi et al also estimate the amount that product development costs have risen by significantly more than inflation in the past: 7.4% per annum above inflation.[69] This increase in development costs may be due to lower success rates for product development, which may be caused by more stringent regulation, or less successful analysis of early drug candidates.[70] The rise in product development costs may also be due to increasing complexity of targets.[71]

5. Bibliography

A conversation with Bhaven Sampat on 05/02/13. (2013). Retrieved from Sampat 05-02-13 (public).pdf

An Advance Market Commitment for AIDS Vaccines: Accelerating the Response from Industry. (2006). Retrieved from

Banzi, R., Moja, L., Pistotti, V., Facchini, A., & Liberati, A. (2011). Conceptual frameworks and empirical approaches used to assess the impact of health research: an overview of reviews. Health Research Policy and Systems / BioMed Central, 9(1), 26. doi:10.1186/1478-4505-9-26

Barder, O. (2010). Break Out the Champagne! The AMC Delivers Vaccines | Center For Global Development. Centre for Global Development website. Retrieved September 13, 2015, from

Berndt, E. R., Glennerster, R., Kremer, M. R., Lee, J., Levine, R., Weizsacker, G., & Williams, H. (2005). Advanced Purchase Commitments for a Malaria Vaccine: Estimating Costs and Effectiveness. National Bureau of Economics Research Working Papers, 11288. Retrieved from

Birkett, A. J., Moorthy, V. S., Loucq, C., Chitnis, C. E., & Kaslow, D. C.(2013). Malaria vaccine R&D in the Decade of Vaccines: breakthroughs, challenges and opportunities. Vaccine, 31 Suppl 2, B233–43. doi:10.1016/j.vaccine.2013.02.040

BL 10 Business plan 2008-2013. (2007). Retrieved from

Burrows, J. N., van Huijsduijnen, R. H., Möhrle, J. J., Oeuvray, C., & Wells, T. N. C.(2013). Designing the next generation of medicines for malaria control and eradication. Malaria Journal, 12(1), 187. doi:10.1186/1475-2875-12-187

Buxton, M., Hanney, S., & Jones, T. (2004). Estimating the economic value to societies of the impact of health research: a critical review. Bulletin of the World Health Organization, 82(10), 733–9. Retrieved from

Cockburn, I., & Henderson, R. (1997). Public-Private Interaction and the Productivity of Pharmaceutical Research. National Bureau of Economics Research Working Papers, 6018. Retrieved from

Cost Effectiveness Tool | Center For Global Development. (n.d.). Centre for Global Development website. Retrieved September 10, 2015, from

Costello, J. (2014). Measuring benefits | Giving What We Can. Giving What We Can Blog. Retrieved September 9, 2015, from

Dalton, M. (2014a). Part 3: Estimating the cost-effectiveness of research into neglected diseases | The Global Priorities Project. Open Philanthropy Project Website. Retrieved September 9, 2015, from

Dalton, M. (2014b). Research Cost-effectiveness spreadsheet. Retrieved from

Dalton, M. (2015). Cost-effectiveness estimate based on GiveWell method. Retrieved from

Danzon, P. M., & Nicholson, S. (2012). The Oxford Handbook of the Economics of the Biopharmaceutical Industry. Oxford University Press. Retrieved from

Deworm the World Initiative, led by Evidence Action | GiveWell. (2014). GiveWell website. Retrieved September 13, 2015, from

DiMasi, J., Hansen, R., & Grabowski, H. (2003). The price of innovation: new estimates of drug development costs. Journal of Health Economics, 22, 151–185. Retrieved from–1301-view-168

Drug development and evaluation for helminths and other neglected tropical diseases. (2009). Retrieved from

Exceptional Returns: The Economic Value of America’s Investment in Medical Research. (2000). Retrieved from

Fact sheet: The RTS,S malaria vaccine candidate (MosquirixTM). (2015). Retrieved from


Farlow, A. (2005). The Global HIV Vaccine Enterprise, Malaria Vaccines, and Purchase Commitments: What is the Fit? Retrieved from

Farlow, A. (2006). The Science, Economics, and Politics, of Malaria Vaccine Policy. Retrieved from

Farlow, A., Light, D., Mahoney, R., & Widdus, R. (2005). Concerns Regarding the Centre for Global Development Report “Making Markets for Vaccines.” Retrieved from

Financing & incentives for neglected disease R&D: Opportunities and challenges. (2011). DNDi Outlook. Retrieved from

Financing neglected disease R&D: principles and options. (2010). Retrieved from paper Innovative financing mechanims.pdf

Forsythe, S. (2011). Vaccine Research and Development Perspective Paper. Retrieved from

Frank, C., & Nason, E. (2009). Health research: measuring the social, health and economic benefits. CMAJ : Canadian Medical Association Journal = Journal de l’Association Medicale Canadienne, 180(5), 528–34. doi:10.1503/cmaj.090016

From pipeline to product: Malaria R&D funding needs into the next decade. (2013). Retrieved from

Gardner, M. J., Hall, N., Fung, E., White, O., Berriman, M., Hyman, R. W., … Barrell, B. (2002). Genome sequence of the human malaria parasite Plasmodium falciparum. Nature, 419(6906), 498–511. doi:10.1038/nature01097

Getz, K. A.(2010). Sizing Up the Clinical Research Market. Applied Clinical Trials website. Retrieved September 9, 2015, from

G-FINDER Report 2014. (2014). Retrieved from GFINDER highlights web.pdf

Glover, M., Buxton, M., Guthrie, S., Hanney, S., Pollitt, A., & Grant, J. (2014). Estimating the returns to UK publicly funded cancer-related research in terms of the net value of improved health outcomes. BMC Medicine, 12(1), 99. doi:10.1186/1741-7015-12-99

Grabowski, H., Vernon, J., & DiMasi, J. A.(2002). Returns on research and development for 1990s new drug introductions. PharmacoEconomics, 20 Suppl 3, 11–29. Retrieved from

Gray, A., Fenn, P., Garau, M., Pritchard, C., Godber, E., & Towse, A. (2006). Donor Investment Choices: Modeling the Value for Money of Investing in Product Development, Public Private Parnerships as Compared to Other Health Care and Non-Health Care Interventions | OHE. Retrieved from

Hall, B. H., Mairesse, J., & Mohnen, P. (2009). Measuring the Returns to R&D. National Bureau of Economics Research Working Papers, (15622). Retrieved from

Health and economic benefits of an accelerated program of research to combat global infectious diseases. (2004). CMAJ : Canadian Medical Association Journal = Journal de l’Association Medicale Canadienne, 171(10), 1203–8. doi:10.1503/cmaj.1041634

Hecht, R., Jamison, D. T., Augenstein, J., Partridge, G., & Thorien, K. (2011). Vaccine Research and Development assessment paper. Retrieved from

Hecht, R., Wilson, P., & Palriwala, A. (2009). Improving health r&d financing for developing countries: a menu of innovative policy options. Health Affairs (Project Hope), 28(4), 974–85. doi:10.1377/hlthaff.28.4.974

Hillebrandt, H. (2015a). An update on the effectiveness of Deworm the World Initiative (DtWI) | Giving What We Can. Giving What We Can website. Retrieved September 9, 2015, from

Hillebrandt, H. (2015b). The most cost-effective health intervention? Killing parasitic worms, malaria mosquitos and HIV with one stone. | Giving What We Can. Giving What We Can Blog. Retrieved September 13, 2015, from

HIV Vaccine Approaches - IAVI - International AIDS Vaccine Initiative. (n.d.). Retrieved September 13, 2015, from

How We Assess Charities | Giving What We Can. (n.d.). Giving What We Can website. Retrieved September 9, 2015, from

Increasing the economic impact of Research Councils. (2006). Retrieved from

Innovative vector control interventions. (2009). Retrieved from

Journal information. (n.d.). PLOS Neglected Tropical Diseases: A Peer-Reviewed Open-Access Journal. Retrieved September 9, 2015, from

Karnofsky, H. (2013a). Meta-research update | The GiveWell Blog. GiveWell Blog. Retrieved September 13, 2015, from

Karnofsky, H. (2013b). Scientific Research Funding | The GiveWell Blog. GiveWell Blog. Retrieved September 9, 2015, from

Karnofsky, H. (2014a). A conversation with Anna Bershteyn. Retrieved from Bershteyn 7-17-2014 (public).pdf

Karnofsky, H. (2014b). A Conversation with Micah Manary. Retrieved from Manary 8-28 and 9-30-14 %28public%29.pdf

Karnofsky, H. (2014c). Cancer Cost-effectiveness Analysis. Retrieved from

Karnofsky, H. (2014d). Returns to Life Sciences Funding | The GiveWell Blog. GiveWell Blog. Retrieved September 10, 2015, from

Karnofsky, H. (2015a). Breakthrough fundamental science | The GiveWell Blog. GiveWell Blog. Retrieved September 9, 2015, from

Karnofsky, H. (2015b). Investigating neglected goals in scientific research | The GiveWell Blog. GiveWell Blog. Retrieved September 9, 2015, from

Karnofsky, H. (2015c). Our updated agenda for science philanthropy | The GiveWell Blog. GiveWell Blog. Retrieved September 9, 2015, from

Karnofsky, H. (2015d). Science policy and infrastructure | The GiveWell Blog. GiveWell Blog. Retrieved September 13, 2015, from

Karnofsky, H. (2015e). The Path to Biomedical Progress | The GiveWell Blog. GiveWell Blog. Retrieved September 9, 2015, from

Keane, J., Kvinikadze, G., O’Hara, J., & Palekar, S. (2006). Assessing the Efficacy of Health Research as a Development Strategy in Poverty Reduction Strategy Papers. Retrieved from

Kinge, J. M., Roxrud, I., Vollset, S. E., Skirbekk, V., & Røttingen, J.-A. (2014). Are the Norwegian health research investments in line with the disease burden? Health Research Policy and Systems / BioMed Central, 12(1), 64. doi:10.1186/1478-4505-12-64

Kollewe, J. (2015). First malaria vaccine given green light by European regulators. The Guardian. Retrieved from

Kremer, M. (2001). Creating Markets for New Vaccines - Part II: Design Issues. In A. B. Jaffe, J. Lerner, & S. Stern (Eds.), Innovation Policy and the Economy, Volume 1 (pp. 73–118). MIT Press. Retrieved from

Kreps, D. (1990). A Course in Microeconomic Theory.

Lengeler, C. (2004). Cochrane Database of Systematic Reviews. The Cochrane database of systematic reviews. Chichester, UK: John Wiley & Sons, Ltd. doi:10.1002/14651858

Levine, R., Kremer, M., & Albright, A. (2005). Making Markets for Vaccines. Retrieved from

Mahmoud, A., Danzon, P. M., Barton, J. H., & Mugerwa, R. D.(2006). Product Development Priorities. In Disease Control Priorities in Developing Countries. 2nd edition. World Bank. Retrieved from

Maskus, K. (2014). Post-2015 Consensus: Science and Technology Assessment, Maskus | Copenhagen Consensus Center. Retrieved from

McIntyre, K. M., Hawkes, I., Waret-Szkuta, A., Morand, S., & Baylis, M. (2011). The H-index as a quantitative indicator of the relative impact of human diseases. PloS One, 6(5), e19558. doi:10.1371/journal.pone.0019558

Medical Research: What’s it worth? (2008). Retrieved from

Moran, M. (2005). New approaches to funding drug R&D for neglected diseases. Retrieved from

Mueller-Langer, F. (2013). Neglected infectious diseases: are push and pull incentive mechanisms suitable for promoting drug development research? Health Economics, Policy, and Law, 8(2), 185–208. doi:10.1017/S1744133112000321

Neglected tropical diseases. (n.d.). WHO Website. World Health Organization. Retrieved September 9, 2015, from

Plotkin, S., Mahmoud, A., & Farrar, J. (2015). Establishing a Global Vaccine-Development Fund — NEJM. New England Journal of Medicine, 373(4), 297–300. Retrieved from

Plotkin, S., & Morrissey, S. (2015). Interview with Dr. Stanley Plotkin on a strategy for stimulating and supporting global vaccine research. New England Journal of Medicine website. Retrieved September 13, 2015, from

Pneumococcal AMC - Gavi, the Vaccine Alliance. (n.d.). Gavi website. Retrieved September 13, 2015, from

Ramamoorthi, R., Graef, K. M., & Dent, J. (2015). Repurposing pharma assets: an accelerated mechanism for strengthening the schistosomiasis drug development pipeline. Future Medicinal Chemistry, 7(6), 727–35. doi:10.4155/fmc.15.26

Ridley, D. B., Grabowski, H. G., & Moe, J. L.(2006). Developing drugs for developing countries. Health Affairs (Project Hope), 25(2), 313–24. doi:10.1377/hlthaff.25.2.313

Roback, K., Dalal, K., & Carlsson, P. (2011). Evaluation of health research: measuring costs and socioeconomic effects. International Journal of Preventive Medicine, 2(4), 203–15. Retrieved from

RTS, S. Clinical trials partnership. (2015). Efficacy and safety of RTS,S/AS01 malaria vaccine with or without a booster dose in infants and children in Africa: final results of a phase 3, individually randomised, controlled trial. The Lancet, 386(9988), 31–45. doi:10.1016/S0140-6736(15)60721-8

Scientific research funding | GiveWell. (n.d.). GiveWell website. Retrieved September 9, 2015, from

Scoble, R., Dickson, K., Hanney, S., & Rodgers, G. J.(2010). Institutional strategies for capturing socio-economic impact of academic research. Journal of Higher Education Policy and Management, 32(5), 499–510. doi:10.1080/1360080X.2010.511122

Setback for Malaria Vaccine: Time for an AMC? | Center For Global Development. (2012). Centre for Global Development website. Retrieved September 13, 2015, from

Sinick, J. (n.d.). Cost-Effectiveness in $/DALY for Deworming Interventions | GiveWell. GiveWell website. Retrieved September 9, 2015, from

Staying the course? Malaria research and development in a time of economic uncertainty. (2011). Retrieved from

Sun, E., Jena, A. B., Lakdawalla, D., Reyes, C., Philipson, T. J., & Goldman, D. (2010). The Contributions of Improved Therapy and Earlier Detection to Cancer Survival Gains, 1988-2000. Forum for Health Economics & Policy, 13(2). Retrieved from

TB Vaccine Research and development: a business case for investment. (2013). Retrieved from

The 10/90 report on Health Research 2001-2002. (2002). Retrieved from

The Vaccine Landscape for Neglected Diseases. (2011). Retrieved from

Thomas, L. (2015). What are Neglected Diseases? News Medical. Retrieved September 9, 2015, from

Towse, A. (2005). A review of IP and non-IP incentives for R&D for Diseases of Poverty. What type of innovation is required, and how can we incentivise the private sector to deliver it? Retrieved from

Tremonti, G. (2005). Background papers to Advanced Market Commitments for vaccines A new tool in the fight against disease and poverty. Retrieved from

Trouiller, P., Olliaro, P., Torreele, E., Orbinski, J., Laing, R., & Ford, N. (2002). Drug development for neglected diseases: a deficient market and a public-health policy failure. Lancet (London, England), 359(9324), 2188–94. doi:10.1016/S0140-6736(02)09096-7

Tuberculosis Vaccines: the case for investment. (2006). Retrieved from,_the_case_for_investment.pdf

Weil, D. (2013). Health and Economic Growth. Retrieved from

WHO. (2003). Investing in Health: a Summary of the Findings of the Commission on Macroeconomics and health. Retrieved from

WHO Scientific Working Group Report on Lymphatic Filariasis. (2005). Retrieved from

WHO. (2015). Pilot implementation of first malaria vaccine recommended by WHO advisory groups. Retrieved from

Worldwide country situation analysis: response to antimicrobial resistance. (2015). Retrieved from

Yazdizadeh, B., Majdzadeh, R., & Salmasian, H. (2010). Systematic review of methods for evaluating healthcare research economic impact. Health Research Policy and Systems / BioMed Central, 8(1), 6. doi:10.1186/1478-4505-8-6

6. Footnotes

  1. See Plotkin & Morrissey, (2015) ↩︎

  2. (Hillebrandt, 2015a, section on ‘The case for scientific evaluation’) ↩︎

  3. The study is also dated: (The 10/90 report on Health Research 2001-2002, 2002) p. xviii ↩︎

  4. Plotkin, Mahmoud, & Farrar, (2015), table on page 298 ↩︎

  5. (“Health and economic benefits of an accelerated program of research to combat global infectious diseases.,” 2004) p. 1206 ↩︎

  6. (The Vaccine Landscape for Neglected Diseases, 2011) p.9 ↩︎

  7. Karnofsky (2014b) ↩︎

  8. (G-FINDER Report 2014, 2014) ↩︎

  9. For a discussion of the product profiles needed, see Burrows, van Huijsduijnen, Möhrle, Oeuvray, & Wells, (2013) ↩︎

  10. Karnofsky (2014a) ↩︎

  11. Towse (2005) p. 27 ↩︎

  12. (Fact sheet: The RTS,S malaria vaccine candidate (MosquirixTM), 2015) ↩︎

  13. (Kollewe, 2015) ↩︎

  14. (WHO, 2015) ↩︎

  15. A scientific view of the difficulties is provided in Birkett, Moorthy, Loucq, Chitnis, & Kaslow, (2013) ↩︎

  16. (Innovative vector control interventions, 2009) p.9 ↩︎

  17. Towse (2005) pp. 25-6 ↩︎

  18. (Staying the course? Malaria research and development in a time of economic uncertainty, 2011) p.3 ↩︎

  19. (From pipeline to product: Malaria R&D funding needs into the next decade, 2013) ↩︎

  20. Karnofsky (2014b) ↩︎

  21. (“HIV Vaccine Approaches - IAVI - International AIDS Vaccine Initiative,” n.d.) ↩︎

  22. Karnofsky (2014a) ↩︎

  23. (Hecht, Jamison, Augenstein, Partridge, & Thorien, 2011) ↩︎

  24. (Hecht, Jamison, Augenstein, Partridge, & Thorien, 2011) ↩︎

  25. Karnofsky (2014a) ↩︎

  26. (Tuberculosis Vaccines: the case for investment, 2006) p. 9 ↩︎

  27. (Tuberculosis Vaccines: the case for investment, 2006) ↩︎

  28. (TB Vaccine Research and development: a business case for investment, 2013) pp. 37-41, (Tuberculosis Vaccines: the case for investment, 2006) p. 6 ↩︎

  29. (Drug development and evaluation for helminths and other neglected tropical diseases, 2009) ↩︎

  30. (Innovative vector control interventions, 2009) p. 9 ↩︎

  31. (WHO Scientific Working Group Report on Lymphatic Filariasis, 2005) ↩︎

  32. (Drug development and evaluation for helminths and other neglected tropical diseases, 2009) p. 8 ↩︎

  33. ‘[R]esearch is needed to further improve intervention tools and provide evidence on cost-effective and appropriate implementation strategies.’ (BL 10 Business plan 2008-2013, 2007, p. 2) ↩︎

  34. Barder (2010) ↩︎

  35. Towse (2005) p.20-22 ↩︎

  36. (Ramamoorthi, Graef, & Dent, 2015) ↩︎

  37. Karnofsky (2015c) ↩︎

  38. Karnofsky (2015b) ↩︎

  39. (G-FINDER Report 2014, 2014) ↩︎

  40. (Danzon & Nicholson, 2012) p.538 ↩︎

  41. (Danzon & Nicholson, 2012) p.327 ↩︎

  42. (Cockburn & Henderson, 1997) ↩︎

  43. (Keane, Kvinikadze, O’Hara, & Palekar, 2006) ↩︎

  44. Karnofsky (2013a), Karnofsky (2015d) ↩︎

  45. (Farlow, Light, Mahoney, & Widdus, 2005) ↩︎

  46. (Danzon & Nicholson, 2012) p. 324, 327 ↩︎

  47. See: At the moment, this is only a proposal. ↩︎

  48. (Danzon & Nicholson, 2012) p. 327 ↩︎

  49. (Financing neglected disease R&D: principles and options, 2010) pp. 3-4, ↩︎

  50. See Barder (2010). The impact of the AMC has not yet been fully evaluated (see the 2015 report at (“Pneumococcal AMC - Gavi, the Vaccine Alliance,” n.d.)). Note that an AMC was also proposed for a malaria vaccine (“Setback for Malaria Vaccine: Time for an AMC? | Center For Global Development,” 2012). ↩︎

  51. Although it is likely that incentivisation is less binary than this first-pass analysis suggests: having a larger AMC will probably encourage firms to invest more resources in production than they otherwise would. ↩︎

  52. As discussed in detail above. See, for instance, Farlow, Light, Mahoney, & Widdus, (2005). Levine, Kremer, & Albright, (2005) estimate that an AMC of $3bn is needed, whilst Farlow (2005) estimates a figure closer to $60bn, illustrating the size of the disagreement. ↩︎

  53. (Danzon & Nicholson, 2012) pp. 323-4 ↩︎

  54. Levine, Kremer, & Albright, (2005) e.g. p. x ↩︎

  55. See Ridley, Grabowski, & Moe, (2006), Moran, (2005). ↩︎

  56. (Mueller-Langer, 2013) ↩︎

  57. For evaluative discussion see Towse (2005) p. 45-56, Moran, (2005), Mueller-Langer, (2013), and (Hecht, Wilson, & Palriwala, 2009). ↩︎

  58. (“Deworm the World Initiative, led by Evidence Action | GiveWell,” 2014), Hillebrandt (2015a) ↩︎

  59. ↩︎

  60. ↩︎

  61. ↩︎

  62. ↩︎

  63. ↩︎

  64. [ ↩︎

  65. (Plotkin, Mahmoud, & Farrar, 2015) ↩︎

  66. (DiMasi, Hansen, & Grabowski, 2003) ↩︎

  67. (“Financing & incentives for neglected disease R&D: Opportunities and challenges,” 2011) ↩︎

  68. (Mahmoud, Danzon, Barton, & Mugerwa, 2006) ↩︎

  69. (DiMasi, Hansen, & Grabowski, 2003). (Mahmoud, Danzon, Barton, & Mugerwa, 2006) and (Danzon & Nicholson, 2012) pp. 22-8 also report rising costs. ↩︎

  70. (Burrows, van Huijsduijnen, Möhrle, Oeuvray, & Wells, 2013) p. 24 ↩︎

  71. (Plotkin & Morrissey, 2015), (Plotkin, Mahmoud, & Farrar, 2015) ↩︎