December 9, 2002
Health For Humanity: How Patent Protection Saves Lives
Robert Goldberg, Ph.D.
What is the effect of innovation on human health? There are a lot of questions about whether patents are important to innovation, or whether we need them at all.
This is not a new debate. In the 1950s, Estes Kefauver held hearings about the monopolistic profits of the pharmaceutical industry. He introduced a bill that required compulsory licensing after three years of market life. Kefauver said to Vannevar Bush, who ran the Office of Scientific Research under Roosevelt and who was the first head of NIH: "Isn't it true that penicillin had no patents? And look at all the lives it saved."
Vannevar Bush said, "Indeed, that's true." But then he pulled a letter out of his pocket from the widow of the inventor of penicillin, Alexander Fleming. His dying wish, she said, was that he had indeed patented his discovery, because then he wouldn't have had to scrounge for 10 years to get money from the government to conduct his experiments. He wouldn’t have had to recycle penicillin from people's urine. Millions of more lives could have been saved.
On that note, I’d like to introduce our speakers. First, Julian Morris, director of the International Policy Network, who has written and compiled Ideal Matter, an excellent book on intellectual property.He’s followed by Professor Frank Lichtenberg, my intellectual rabbi on many of these issues, and the Courtney Brown Professor of Business at Columbia University Business School. Then Jamie Love, director of the Consumer Project on Technology, who has probably done more than anyone to focus the issue of patents and research and development costs and their relationship to health-technology access. Finally, we’ll hear from Amir Attaran, research fellow with the Carr Center for Human Rights Policy at Harvard, who has done seminal work on the relationship between patent protection and access to medicines in Africa and other developing countries.
Panel 1: Medical Progress and Public Health: A Look at the Evidence
This morning I'm going to talk about three main ideas. First, the relationship between wealth and health. I'll argue that the two are strongly related, and that, to a large extent, wealth begets health. Secondly, both wealth and health depend on the institutions of a free society: property rights, free markets, and good governance. Finally, I'll argue that the stronger the system of property rights, the more likely the development of health technology.
First, that wealth begets health. If you look over history, there's a strong correlation between the two. In 1870, the GDP of the U.K. and the U.S. was around U.S. $2,500. In the same period, the life expectancy in those countries was about 43. By 1990, GDP in those countries had risen to nearly U.S. $14,000 in the U.K. and over U.S. $18,000 in the U.S. Life expectancy had risen to 76. That's nearly a doubling life expectancy. In large part, I think, the increase in wealth is responsible.
If you compare countries, there is broadly some correlation. For example, if you look at countries where the GDP is still less than half of what it was in the U.S. and U.K. in 1870 -- $1,000 a year in Malawi, for instance, life expectancy is 40 years or less, lower than the life expectancy was in the U.S. and the U.K. in 1870.
There are anomalies. For example, India, where per capita income is, depending on how you measure it, between $1,000 and $2,000, the life expectancy is about 61, which is reasonably high. In China, where per capita income again is about U.S.$2,000, life expectancy is about 70. But I will explain later why that is.
If you go back to the observation of this improvement in health over time, in what are now rich countries, it's clear that the increase in wealth enables such things as increased sanitation, the draining of the towns (removing unhealthy sewage), and the ability for people to buy soap and other things that remove dirt and thereby reduce their exposure to bacterial infections. Sanitation is a very important part of the battle to reduce the incidence of disease, especially among children.
Wealth also enables people to invest in the development of new technologies, such as penicillin and the Salk vaccine. Investments in such technologies were probably lower than they might have been because of the lack of protection of human capital. But the fact that we were wealthy meant that we could invest in the development of this technology. And the effects of early medical innovations were also relatively more dramatic in part because they were the first and were directed at the worst problems.
Likewise, the draining of swamps and the use of chemicals such as DDT enabled us to eradicate malaria in the US and Europe. Even in the time even of Thomas Jefferson, Washington, D.C. was a malarial area and for that reason was avoided during the summer months.
But the question remains how to generate the national wealth that allows these improvements in the nation’s health. If wealth is important to health, the question is: “how does a nation become wealthy?”
The best recent explanation of why rich countries become rich has been provided by Peruvian economist Hernando de Soto. In his wonderful book, The Mystery of Capital, he explains the role of property rights in enabling people to have both the incentives and the ability to generate wealth. The basic premise is as follows:
If a person owns a piece of land, she knows that she can benefit from the stream of income that land generates. With that knowledge comes an incentive to invest in improvements that are expected to increase both the revenue stream and the value of that land. But if she doesn't own the land, the opposite is the case. She won't have the ability or the incentive to generate revenue. So real property, de Soto argues, is an absolutely essential element in economic development.
De Soto has studied the presence or absence of formal property rights in many countries. He observes that huge sums have been invested in improvements to land even though those investments are not realizable because of the lack of formal ownership. In many countries people have built houses and other buildings even though they have no way of legitimately transferring them. As a result, they cannot be used as ‘capital’ – they cannot be mortgaged. De Soto estimates that the total value of this ‘dead capital’ at $9 trillion. But this is probably an underestimate of the total because there are many countries that De Soto hasn't surveyed, for example, India, where huge amounts of capital are missing for exactly the same reasons.
But property rights alone aren't the only requirement for generating wealth. For an individual to be able to benefit from any property, he or she must be able to sell it. In order to sell it, you need to have markets. Those markets are dependent on the right of individuals to form legally binding contracts. If people can contract with each other freely, then they can exchange goods. And if people can exchange goods, they can specialize in the production of certain goods – one person can produce wheat and another bread. As Adam Smith observed, the division of labor that results from specialization leads to increases in efficiency. It also spurs innovation, as suppliers compete for the market, developing more desirable products and looking for more efficient ways of producing goods. The result is generalized economic growth.
Conversely, most countries that lack property rights and markets – or have insufficient protection of property rights and over-regulated markets – are typically also poor. So there is, I think, ample evidence to suggest that this model of capitalist development is the best for humanity.
Having said that, as I observed earlier, although there is some correlation across time between wealth and health, the correlation across countries at any point in time is not necessarily as good. One reason is that technologies innovated and developed in capitalist countries have been transferred to non-capitalist (or less capitalist) countries.
Even so, in poor, badly governed countries, the ability to benefit from these technologies is limited. In India, for example, millions of people die every year as a result of inadequate sanitation and a lack of clean drinking water. The main reason for this tragic state of affairs is that the supply of drinking water in most parts of India is controlled by the state. However, in those places where ownership of water has been decentralized (so the local community or a private company owns and manages the supply), people generally have better access to drinking water. The reason is simple: where people own a resource, they have an incentive to invest in improving that resource. This applies to water as much as it applies to land.
The same applies to the provision of healthcare. In India, there are abundant inexpensive medicines available from the approximately 20,000 pharmaceutical companies. Nearly all these medicines were developed elsewhere. Since 1970, India has had no patent protection for pharmaceutical products and many of the products manufactured in India are still on patent elsewhere.
But, according to the WHO, only 30 percent of the Indian population has access to these medicines. There are various reasons for this, but the most important one is the lack of wealth. Per capita income in India is extremely low, especially in rural areas. Another important reason is the lack of health infrastructure, which is a consequence of failed attempts by Indian states to supply health. This crowds out the private sector and undermines access.
I spent a few months in India this year and there met with some people in the health community. I was surprised to hear that many people in India look to the national health system in the U.K. as an ideal model. With straight faces they compare a rich, mostly capitalist country – which has a largely dysfunctional nationalized health system (I speak from personal experience), but nevertheless a system functional enough to distribute medicines to the population – with a country whose per capita income is about one-twentieth that of the U.K. and clearly has an even more corrupt, underfunded, and inefficient health-infrastructure.
The situation in sub-Saharan Africa is even worse, not only because of the lower per capita income of many sub-Saharan African countries, but also because of the disastrous systems of governance that have arisen in those countries over the past 20 years. I visited Zimbabwe in 1995, when it was barely functioning following 15 years of misrule of Robert Mugabe. One of the colleagues with whom I was traveling had been there in 1990 and observed that conditions had significantly worsened. The statistics tell the tale. From 1990 to 2000, Zimbabwe's GDP fell by about 50 percent as a result of incompetence, corruption, and widespread threats to the enforceability of property rights and contracts. That has undermined entrepreneurial activity, leading to massive capital flight and all sorts of other adverse consequences. For example, rates of malaria and AIDS both rose dramatically.
One may contrast this with the situation in Uganda, where after years of misrule by Idi Amin and others, President Musaveni has instilled some semblance of order, reducing corruption and building confidence in the market system. He has also instituted a highly effective campaign to tackle AIDS. The result has been that since 1992, rates of HIV infection have fallen by more than 50 per cent.
So, to sum up, improving health in poor countries requires a combination of improved protection of property rights – and I think that also includes protection of intellectual property rights. (In countries with a significant pharmaceutical industry, the protection of patents on pharmaceutical products and diagnostics will encourage the development of new diagnostics and treatments.) It also entails free markets, bound by effective enforcement of contracts. And it entails good governance.
In combination, these things will result in technological innovation. They will enable investments in sanitation and clean water, which are essential. And they will enable investments in education and in drug development; both absolutely essential in dealing with the various scourges that afflict underdeveloped countries.
The research that I've been engaged in, for the last 8 or so years, indicates that new drugs confer several important benefits: longer life, fewer limitations on work and other activities--also known as increased quality of life--and reduced overall medical expenditures.
I will only talk about the first two benefits today. Then I'd like to go on to talk about a study that I've done that indicates that actually people in the poorest initial health receive the most benefit from new drugs. Finally, I’ll argue that public policy has an important impact on the development and launch of new drugs.
So first, the effect of new drugs on longevity. The hypothesis that I tried to examine in a number of different studies is that mortality from a condition or a disease is inversely related to the number of drugs available to treat the condition. So over time, as new drugs are developed to treat given conditions, mortality from those conditions declines, and also the quality of life and other attributes also tend to improve.
I'm going to talk a little bit about four kinds of evidence. First, about two specific diseases in the United States, breast cancer and HIV. Breast cancer kills about 40,000 women per year in the United States. But there has been a huge increase in the number of drugs available to treat breast cancer. Beginning in 1987, there was a steady flow of about one breast cancer drug per year approved. And then, from 1995 to 1999, 12 new breast cancer drugs were approved.
Based on Centers for Disease Control data, the breast cancer death rate, age adjusted, has a very dramatic pattern. Breast cancer mortality was rising gradually in the early 1980s, then it started to decline around 1990, and the decline accelerated in the mid-1990s. There's an extremely strong correlation. If one does statistical analysis of the data to examine the correlation between mortality from breast cancer and the number of drugs approved, that confirms a very strong correlation. In fact, it suggests that the average new drug for breast cancer reduces the number of breast cancer deaths in the United States by about 500 per year.
Now, the mean age of death from breast cancer is about 10 years lower than mean age of death of women from other causes. In other words, women who die of breast cancer die at about the age of 68, on average. Women who die from other causes die at about 78. In that sense, breast cancer robs women of 10 years of life, on average.
That implies that the average new drug for breast cancer reduces the number of life years lost by about 5,000 per year. So once a new breast cancer drug is approved by the FDA, in that year and all subsequent years, 5,000 fewer life years will be lost to breast cancer in the United States alone.
Now, I've done this kind of analysis also for HIV. There's a similarly very strong pattern, a relationship between the number of HIV drugs approved and the reduction in the number of HIV deaths in the following year. And there was a very dramatic reduction in HIV mortality in the mid-1990s. That tracks very closely with the explosion in HIV antiretroviral drugs in or around 1995. In fact, I estimate that the annual number of HIV deaths has been reduced in the United States by about 6,100 per additional HIV drug approval.
Thus, the deaths prevented per drug are about 10 times as high as it is for breast cancer. Because people who die from HIV tend to die more around age 40 rather than age 70, the number of life years saved by the average HIV drug approval is in fact very, very high.
So that’s evidence about two diseases, breast cancer and HIV. But I've also examined the relationship between mortality and new drug approvals for all diseases. I estimate that the new drugs introduced in the United States during the period 1979 to 1998 increased the longevity of Americans by at least 5 months during that period. In other words, the average person who was born in 1998 is going to live 5 months longer than the average person born in 1979 because of the drugs that were approved during that 20-year period.
I also have some preliminary evidence about international drug launches, and that evidence suggests that new drugs launched internationally during the period 1987 to 2000 increased longevity by about 9 months globally during that period. So there's 9 months of added life expectancy on a worldwide basis due to the new drugs that were launched internationally.
A lot of drugs are never launched in certain countries, and the longevity gain could be even higher if there were greater diffusion of new drugs. But with the actual diffusion of new drugs, that's the longevity gain that I estimate.
Other economists have shown that the value to people of increases in longevity is very high. People are willing to pay a lot in order to live an extra year. So the economic value of the longevity gained is quite significant.
The second hypothesized benefit of new drugs is fewer limitations on work and other activities. I've done a study that examines the relationship between (a) the change in the percent of people with a condition who are unable to work, because they have that condition and (b) the increase in the number of drugs available for that condition. Data on the former were provided by the National Health Interview Survey, which asks people year after year: "Do you have diabetes? Do you have asthma? Do you have hypertension?" And if they say, "Yeah, I have that," then they say, "Are you unable to work because of that condition?" Accordingly we can follow different diseases over time, and we can estimate the percent of people with given conditions that report being completely unable to work because of those conditions.
I looked at those data and at the relationship between the change in inability to work and the increase in the number of drugs. The greater the increase in the number of drugs available to treat a condition, the greater the decrease in the probability of being unable to work due to that condition. So that suggests that new drugs increase productivity as well as longevity.
In fact, if we take all the drugs that were approved during the period 1983 to 1996, I estimate that the new drug approvals during that period reduced the number of people who were unable to work in 1996 by about 1.4 million people. If we estimate the value of the ability to work at the average wage of about $30,000 per year, then the benefit of that is about $43 billion per year.
Moreover, people who are able to work had reduced work-loss days and lower absenteeism. In fact, we can factor in about 100 million fewer missed workdays by people who do have jobs, and the value of that may be about $10 billion a year.
My second key point is that people in the poorest initial health receive the most benefit from new drugs. With certain kinds of technological innovations, like computers, the evidence shows that computers are good for productivity and economic growth. But they have a bad side effect, in that they tend to increase inequality, because the most educated people benefit from computers a lot, and the least educated tend to be left behind by information technologies. So IT has probably increased inequality.
However, drugs, I think, are very different. And in order to look at this, I (along with a colleague, Suchin Virabhak) used data from the 1997 Medical Expenditure Panel Survey, a government survey that covered about 35,000 people and all of the 240,000 medicines that they took. There was both an initial interview of each person, and then a subsequent interview, about a year later. During that time the person may have taken different drugs, and we know what drugs a person took. And in particular, we know how new the drugs were; we know the FDA approval year of the drugs that the person took during that period.
We find that people in the poorest initial health receive the most benefit from new drugs. That is, if someone is in initially good health, their post-treatment health will be improved by taking newer drugs, but it won't improve that much. If someone is initially in very bad health, however, new drugs matter a great deal, and the person's post-treatment health would be strongly related to the vintage of the drug that that person is taking. That's very consistent with the hypothesis stated there.
My last major point: Public policy has an important impact in the development and launch of new drugs. I have a couple of pieces of evidence to support that. One comes from the Orphan Drug Act of 1983. The objective of the Orphan Drug Act was to promote the development of products that demonstrate promise for the diagnosis and/or treatment of rare diseases. Congress subsequently defined a rare disease as any disease that affects fewer than 200,000 Americans. And the Orphan Drug Act had a number of different provisions providing incentives for the development of drugs for rare diseases, including 7 years of exclusive marketing, a tax credit and research grants.
So Congress decided we really want to try to encourage the industry to develop drugs for the treatment of rare diseases. This provides a nice policy experiment. What were the impacts of the Orphan Drug Act?
I've done some work with Joel Waldfogel of the University of Pennsylvania to analyze this. One thing that we can know--and this is purely from the FDA data--is that the Orphan Drug Act led to about a twelve-fold increase in the rate of innovation for rare diseases. The average annual number of drugs for rare diseases brought to market went from one in the decade before the Orphan Drug Act to 12 in the following 15 years. So there was a huge increase in the rate of innovation for rare diseases. That's directly from FDA data.
Did this actually improve the health outcomes of people with rare diseases? One way to measure health outcomes is to look at mortality and, in particular, to look at mean age of death. And so, using mortality data, we calculate the age of death, according to how common the cause of death is.
The most common form of death is heart attack. Lung cancer is also a relatively common form of death. In 1980, the mean age of death from the most common diseases was about 73. The mean age of death from the least common diseases, meanwhile, was about 50.
So people tend to die young from rare diseases. In part, that's because there aren't many drugs for rare diseases. As the prevalence of a disease increases, the mean age of death from that disease increases.
However, the good news for people with rare diseases is that between 1980 (before the Orphan Drug Act) and 1995 (after the Act), the mean age of death increased. It increased especially for rare diseases. That is, people with relatively rare diseases saw the largest increase in longevity, or mean age of death.
People who had very common diseases, who died of heart attacks, died 2 years later in 1995 than they did in 1980. But people with rare diseases increased their longevity by about 7 years. So there was a much greater increase in longevity among people with rare diseases than among people with common diseases. That's consistent with the idea that the Orphan Drug Act increased the longevity of people involved in that category.
My last point relates to the role of public policy. I've looked at the impact of the government role in the pharmaceutical market on international drug-launch behavior. For example, if a drug has been launched anywhere in the world, there's a 39 percent chance that it has been launched in the United States after 2 years. And there's a 46 percent chance that it's been launched after 4 years. In contrast, in Canada, the probability of a drug launch is much lower. After 2 years, it's only half as likely that a drug had been launched in Canada than it is that it's been launched in the United States. So we can measure the probability of drug launch, which is a measure of the availability of new drugs in different countries.
I've hypothesized that the probability of drug launch in a country after a given amount of time depends on public policy in that country, among other things. One measure of public policy is, what percent of total purchases of pharmaceuticals in a country are purchased by the government?
In the United States, only 13 percent of pharmaceuticals are purchased by the government. But in the United Kingdom, 65 percent of drugs are purchased by the government. In Luxembourg, 83 percent of drugs are purchased by the government. And in fact, there's an inverse correlation between (a) the public share of pharmaceutical expenditure, and (b) a 10-year launch probability. In other words, in countries where the government is a larger player in the pharmaceutical market, where it is buying most of the drugs, there's a lower probability that drugs will be launched there.
There may be two reasons for this. First, the government often refuses to purchase the newest drugs. The Veterans Administration will not put any drug on the VA formulary unless it's been approved for at least a year. So it sort of automatically says, any drug that's less than a year old is not going to be reimbursed by the VA unless there's a special exception. Second, the government tends to pay lower fees than private payers, which discourages the launch of new drugs.
To conclude, I have made three points. First, new drugs confer several important benefits, among them longer life and higher productivity. Second, the evidence indicates that people in the poorest initial health receive the most benefit from new drugs. Third, public policy has an important impact on the development and launch of new drugs.
Panel 2: Do Patents Prevent Access?
I think that everyone that's working on this issue is aware of how many different problems there are; I'm just going to be describing one of the problems here.
High prices definitely are a barrier to access. Consider this important case. The United States government forced Thailand to impose a non-patent, regulatory exclusivity on quite a few products. The drug fluconazole, which is not under patent in Thailand, was priced at around 200 baht per pill, the local currency. This drug is used primarily in the United States for vaginal or toenail fungus. The primary interest in Thailand was for the treatment of cryptococcal meningitis, from which about 10 percent of the late-stage AIDS patients were suffering. It's a disease that results in splitting headaches, blindness and deafness.
When the NGOs on the ground in 1998 got the exclusivity removed, the price fell to 6.5 baht in 9 months as companies entered the market. This particular case sent a shock through the AIDS community, because they saw for the first time the link between the IPR regime, competition and access, in a way that made them understand that this was an issue they had to focus on.
Or consider what happened in Brazil. Brazil started issuing patents in 1997, so all the AIDS drugs that were invented in '96 and before were off-patent there.
Brazil initiated a program of universal access to AIDS drugs. But when they first started to buy generic products, the prices were relatively high because there wasn't much of a market. Brazil was the first country where the government got involved and really decided to create a market. They spent about U.S.$150 million per year on imported generic drugs. That was enough to create a competitive market of Asian producers providing fine materials and ingredients from India and China and other places.
This decision by Brazil to buy generic drugs made the current cheap prices in Africa possible. Brazil, which was a middle-income country, created a competitive market for generics. The floor dropped out of the fine materials market, and now you can get $250 anti-AIDS cocktails in Africa.
But we have access problems, and that's what everybody starts out with. And not just HIV. Singular, which my son takes for asthma, is not on formulary in Brazil because it's too expensive. Everyone likes to see a lot of people get the drug, but only at the right price, right? So there's a conflict.
So here's how we fund drugs. What we do is we say, "You have an invention. You have a 15-year monopoly on the product," or whatever the period is, depending on how many extensions and sort of games you can play in the patent system. You just go market that drug like crazy, and you get every doctor to prescribe it. But if you depend upon higher and higher prices to fund R&D, then a lot of people just aren't getting access to drugs. So we're trying to think about how we fund R&D without denying patients access.
The U.K. Commission on Intellectual Property Rights, the Federal Trade Commission, the Department of Justice under the Bush administration, are all grappling with this question. They're brainstorming. They're asking themselves: What do we really think about intellectual property rights?
The Internet has changed our perspective. Private property rights and intellectual property were essential to the development of civilization as we know it. But Internet protocols developed basically on volunteer and public support, with no intellectual property rights, no patents, no copyrights. HTML was designed with that little feature for “view source.” You can immediately see what somebody else did, borrow it, modify it, take it apart, and use it to create your own thing. And so, almost overnight, the Internet became the biggest source of innovation and publishing success you can possibly imagine. It just radically transformed society.
So a lot of people who looked at the Internet, who had believed property rights were vital to commerce, said to themselves: I don't get it. It was like a blizzard in Rio. It's not supposed to happen, right?
The key is to create a mechanism to offset reduced standards from weaker IP regimes so that we can provide people in poor countries with access to drugs. We don't have to end R&D. We can actually use treaty mechanisms to pump up the R&D, under our control. We can create a treaty that creates minimum national contributions to R&D and transparent investment flows.
We can allow countries to fund the R&D anywhere they want. They can do high prices, strong IPR, the U.S. model. They can do public funding, also the U.S. model. They can do research mandate. They can go completely open source, no IPR, and have every product be a generic drug.
Or we could create treaties directly on parallel trade and reference pricing, allowing there to be a low price in the developing world and a high price in the U.S. So if an American company puts a drug on the market in Africa at a lower price, it doesn’t feed back into the Veterans Administration reference pricing scheme, or its way back to the U.S. through parallel trade, allowing the company to maintain it’s margins in the U.S. market.
I'm going to try to approach this problem from the following perspective: What can we do to improve the health of the very poorest people in the world, and how much do patents have to do with that stated mission?
In the United States between 1996 and 1999 there was a 70 percent decline in AIDS deaths occasioned by the use of antiretroviral drugs. In Africa, where those medicines are not available, and prevention services are poor, and the prevailing strains of HIV and their genetics are more aggressive, this is not the case. Rather, the reverse is true, and AIDS mortality is soaring. It is heading stratospherically high. The differences between us, living in the U.S., and Africans or Indians are so tremendous as to really baffle our imagination. You cannot believe it until you see it, it is so tremendously devastating.
So, obviously, we would like to find a way to help. But given limited resources and political capital, how can we best achieve that help? How can we best focus our efforts?
Supposing we do want to provide medicines to people in developing countries, including antiretrovirals, as I think we must, we have to understand why those medicines aren’t available now.
The first necessary step to getting medicine to poor people is, obviously, to make it affordable, and there are currently two ways to do that for the AIDS medicines and some others. One way is that many of the brand-name pharmaceutical companies have significantly discounted their products in developing countries. The other way is, of course, through competition. Rather than relying on a corporate policy decision about discounting prices, you can bring prices down through generic competition, which is normally possible only where there are no patents. Competition is not possible where patents do exist, unless you seek to voluntarily or compulsorily override those patents, using a voluntary or compulsory license.
Either of these two methods can get you to a point of having more affordable medicines. They're parallel strategies; they're not mutually exclusive. You then need to make sure those medicines meet the regulatory requirements of the country where you wish to use them. They need to be registered with whatever the equivalent of the FDA is in Zambia. In each country, national authorities of the right sort need to be apprised of and consent to the use of the medicine in their country.
Then you need money to purchase the medicines, primarily through international aid funding. I cannot drive this point home enough: Someone has to pay, and in developing countries, the economies themselves are much too weak to undertake that burden on their own. It is an impossibility.
Finally, once you actually have the medicines, you need the capacity to deliver them. That means having a certain number of trained clinicians, nurses, diagnostic facilities, and the social setting in which people want to come forward for treatment. With AIDS, for instance, there's enormous stigma around the disease and people are actually afraid to be diagnosed, because the stigma of a positive diagnosis would be crushing in many cases.
I was recently in Malawi and I witnessed a person quite literally die in the clinic where they were waiting to be tested for HIV. They came when they were so ill, they actually died while they were waiting their turn to be tested. The stigma prevents people coming forward to seek help when they could be saved; they only come when they are very ill and near death. All these economic, biomedical and even social factors have to be addressed.
Now one hypothesis about why the medicine delivery chain I’ve just described doesn't work is that patents knock out the ability to pursue generic competition. The theory is that if patents prevent the entry of generic medicines, then patent-holders are free to set their prices at a high level, rendering the medicines unaffordable, stopping the delivery chain at the very first step.
The alternative hypothesis, which I favor, is that it's a lack of international aid financing that prevents progress along the delivery chain. There is good evidence for this. I did a study in 2001 with Lee Gillespie White at the International Intellectual Property Institute, surprisingly the first on the subject, that asked the basic question of how often AIDS medicines are patented in Africa.
Why did we ask that? Well, if you're theorizing that patents stand in the way of access, then naturally you should be able to show that, indeed, the medicines are patented. That theory doesn't hold if the medicines aren't patented. Patents that don't exist cannot be barriers to treatment, if you follow my logic.
So let me explain some basic facts about patent law. There is no such thing as a global patent, and patents are obtained country-by-country or sometimes region-by-region. It's completely possible for a medicine to be patented in Canada and the U.S. and Germany, and not be patented in Peru and Swaziland and Laos, for example. This is why one has to study patent status on a per-country basis.
So, in our 2001 study we investigated the patent status of each of the 15 antiretroviral medicines that then existed for AIDS in each of the 53 African countries. We discovered, to our surprise, that most the time the drugs weren’t patented, except in a few countries. While individual cases vary very widely, the typical drug among these 15 antiretrovirals was patented in only 3 nations in Africa, and not 53 nations as we might have found.
Because this was a surprising result, we then narrowed our focus to just the 40 African countries in which we found at least one medicine under patent, in order to ensure the lack of patents wasn’t merely the result of a lack of suitable patent laws. Even among those 40 nations which were undoubtedly capable of having patents for all 15 AIDS drugs, however, we found that typically only 4 were actually patented. What this means is that patent laws are not being used to their fullest in the low-income countries and middle-income countries of Africa, and that even where the option to seek a patent exists, frequently the option is not exercised.
Overall, in the 792 cases where a drug could theoretically be patented (or 15 drugs times 53 countries) there were roughly 170 patents issued. So patent coverage, far from being ubiquitous, only exists about 21 percent of the time.
Of course, this is a generalization, and there are exceptions. South Africa is the African leader in patent coverage, with 13 of the 15 medicines under patent. In South Africa one could make the argument tentatively that patents might be standing in the way of access. But it would require serious investigation to determine if this were really true, because other factors could stand in the way also. For example, in South Africa, the African National Congress government rejects the use of antiretrovirals, and has for years. President Thabo Mbeki has said he does not believe HIV is the cause of AIDS. The government even had to be sued by its own citizens and ordered by the court to furnish one antiretroviral, called nevirapine. With so much bad faith at work in South Africa, it’s a tall order to say that patents are the cause of there being little AIDS treatment.
Our data also provide an opportunity for something of a thought experiment. If patents really do stand in the way of AIDS treatment in Africa, we should expect very little treatment to be occurring in South Africa where nearly all the antiretrovirals are patented, and quite a lot of treatment to be occurring in Namibia or Mozambique where none of the antiretrovirals are patented. Both countries are comparably wealthy; Namibia is a bit poorer, but not all that much, whereas Mozambique is much poorer. In short, there should be an inverse relationship between the number of patents and the proportion of AIDS patients receiving treatment.
However, no such relationship is evident. In fact, according to the WHO, the World Health Organization, across Africa, only 30,000 people out of between 3 and 4 million who need treatment are getting it. That's 0.1 percent. Basically, nobody is getting treated anywhere in Africa, irrespective of whether they're in a country with an abundance of patents or a dearth of patents. So logically something else must be preventing the 99.9 percent of people who need medicines from getting them.
This makes for a compelling analysis, I think. Nevertheless, since AIDS drugs are taken in combination rather than individually, it is theoretically possible that a patent on a single medicine could prevent patients from getting the full combination treatment regimen. We therefore studied this too, and we found that some clinically outstanding treatment regimens are nearly totally unpatented throughout Africa, whereas other combinations are heavily patented. For these latter regimens, and where patents exist, the patents are in fact potential barriers to treatment.
Now, what happens if that possibility occurs and, for instance, a patented medicine for which there is no alternative is not sold at a fair discount by its inventor? What do you do then, because the situation is obviously unacceptable? In that case there's an ethical imperative to take action. Public health has to take priority over patents.
There are two approaches one can take in such a case: a voluntary or a compulsory approach. Using a voluntary approach you might ask the patent-holder to waive its patent rights or to license them to somebody else. And if that fails, you could adopt an approach such as compulsory licensing, which is roughly speaking to expropriate the patent from the patent-holder and giving it to somebody else.
Certainly the proper policy order is to take the voluntary measures before going on to the compulsory ones, exactly as if you owned a patch of land and the government wanted to put a road through your land. They would first try to buy the land off of you voluntarily, by negotiating a selling price. If you were totally unreasonable and refused to sell, then they would expropriate it.
Very few cases like this actually exist, however. Pricing and availability data from Medicins Sans Frontieres shows that often the branded product compares very favorably to a high-quality generic one (one approved by the World Health Organization). There is not, at this point, a hard-and-fast rule about whether brand-name discounted products are less expensive than the high-quality generics or vice versa. In some cases, there simply isn’t any high-quality generic. And in other cases, surprisingly, the generic is more expensive. It depends from case to case.
These data, however, have not ended the debate over patents standing in the way of treatment access. So let me try another thought experiment: what would be the outcome if we could magically eliminate every single patent worldwide, enabling unlimited generic competition? That experiment has been effectively carried out in India. India currently has no AIDS medicines under patent. India produces more generics for the Third World market than anywhere else in the world. The drugs are locally manufactured, and priced in rupees, not dollars. And despite all this, and despite the fact that India has more AIDS patients than anywhere else on Earth, the WHO estimates that 20,000 Indians out of roughly 10 million with HIV are getting treatment. That, again, rounds out dismally to about zero percent.
So in this setting, where there are no patents, where there's ample generic manufacturing capacity, where there are AIDS patients as deep in the queue as you could imagine in desperate need of treatment, patients still aren’t being treated.
Which brings us back to the alternative hypothesis about the breakdown in the medicine supply chain: international aid. There has been, to put it bluntly, an enormous political failure in our lifetime to deal with AIDS, malaria, TB and other diseases. It is going to be the single-most embarrassing thing about our generation when historians 100 years from now measure us and assess us. It will be shocking. It will make the Holocaust and great wars look, in numbers at least, like footnotes compared to the disease-induced genocide.
And frankly, it happened because governments were sleeping at the wheel. Our politicians blew it. They failed. Look at what the G7 leaders said about AIDS at their annual summit in 1987 [slide]. They said that AIDS was one of the biggest health problems in the world and international efforts needed to be made to stem the pandemic. But nothing was done.
In 1999, the total amount of money spent by the world’s 23 richest countries and the World Bank to deal with AIDS in Africa was $125 million. That’s enough to pay for five miles of freeway in California. It is somewhat over half the production budget of the movie "Titanic." Divided by Africa’s 25 million people, it’s $5 per person per year. That is how much we, acting collectively as the developed and rich world, cared about this pandemic.
With that neglect, AIDS grew to its current proportions for the very simple reason that we did not spend the money or make the political effort to counter it. Moreover, the situation has not much improved. In 2000, the U.S. pledged a maximum of $112 million for AIDS for the 2.4 billion people in low-income countries, and we almost certainly actually spent less. That economic powerhouse, Germany, scraped together $3 million that year for the same purpose.
This is all part of the declining trend in donor assistance. It’s not a uniquely Republican or Democratic failing. Ever since the 1970s, the amount of foreign assistance that the U.S. gives to developing countries has been a pittance. Total assistance is currently about $1 per person per year for the neediest countries on Earth.
There are weapons of mass salvation in the world, and they include medicines, and we are not deploying them. Each of the 23 donor economies mentioned above pledged to give 0.7 percent of their GNP as foreign assistance. Only five of them currently meet that target. And the United States, at 0.1 percent of its wealth dedicated to foreign aid, is the least generous donor economy in the world. Even Greece is more generous than we are.
Where is this going to take us? In places like Botswana, currently, 40 percent of adults have HIV. Four out of 10 adults are going to be dead in 10 years. They're the teachers. They're the police. They're the religious leaders. They're the parents. They're the authority figures of all kinds. They're the ones who underpin the society. They're the ones who teach the kids how to play music and how to read and write. They teach them the traditions.
In ten years, they're dead. They're gone.
Botswana will become a society with no mentors, no authority figures, where children and teenagers are the most numerous people, and sit on top of the world's most productive diamond mines. Throw in some guns, and what's going to happen? This is the "Lord of the Flies" comes to life.
And this scenario is going to replicate itself across Africa, as it's been doing for two decades while the epidemiologists and virologists have screamed, and the politicians do nothing impressive about it.
So in the end, if you really want to assist in developing countries, the issues of patents can be significant. I do not wish to be understood to say that they are they of zero significance; they are not. They are important; they need to be addressed. But it is the thinnest of icing on the cake. Finance, more than anything else, is the limitation right now. Thank you for listening.