Are doctors and pharmaceutical companies to blame for the opioid epidemic? On this episode of the Capitalisn’t podcast, hosts Kate Waldock and Luigi Zingales look at the role of supply and demand in fueling the distribution of prescription painkillers, and discuss the regulatory ramifications for medical marijuana.
Speaker 1: Opioid abuse in the United States is at epidemic levels.
Speaker 2: This is probably the worst drug situation in our country in decades, if not a century.
Kate: Hi, I’m Kate Waldock, and I’m a professor at Georgetown University.
Luigi: And I’m Luigi Zingales, and I’m at a professor at the University of Chicago.
Kate: You’re listening to Capitalisn’t. A podcast about what’s working in capitalism today ...
Luigi: … and most importantly, what isn’t.
Speaker 1: The impact of the opioid epidemic is stunning, with 300,000 American deaths since the year 2000.
Speaker 5: Despite decades of advancements in healthcare, diet, and safety, middle-aged white Americans are now living shorter, not longer, lives.
Speaker 1: Oxycontin’s maker, Purdue Pharma, announcing it is cutting its sales staff in half. And starting today will no longer have reps visiting doctors’ offices to discuss opioids.
Kate: On today’s episode we’re going to be talking about the opioid crisis. I am positive that this is not the first time that you’re hearing about this. It’s been all over the news. In fact, maybe you’re sick of hearing about it, but we’re going to try to look at this issue through the lens of economics. Through the lens of supply and demand.
On the demand side, was it the case that people just wanted a bunch of opioids, and that’s what drove the crisis, or on the supply side was it the case that doctors and pharmaceutical companies were pushing opioids onto people? There’s some evidence that there could be a link between these painkillers and the use of illicit opiates like heroin and Fentanyl, which are also very dangerous, but we’re focusing more on the prescription drugs.
Luigi: So just to be clear, the demand story is not just people take opioids as a recreational drug. Some they might, but I think that the aspect we want to understand is to what extent this a result of economic despair.
Kate: We want to start today’s episode by going back to October of 2015, when Angus Deaton, a professor at Princeton University who was renowned for his research, his economic research about poverty and health, won the Nobel Prize in economics. And in the same month that he won the Nobel Prize, a groundbreaking paper written by him and Anne Case, also a professor at Princeton, was released about how opioids were on the rise and the people who were suffering from opioid deaths were primarily concentrated amongst white, non-Hispanic people and people who were in the middle class.
Luigi: The way I remember the paper is that he shows that for the first time in basically a century the life expectancy of white people was not going up. The life expectancy of most groups was going up, but the life expectancy of white males was not going up and the reason was that many of them were dying of overdose or suicide and this was mostly concentrated in the 40–55 group.
Kate: That and it was in people who were suffering from these deaths, opioid-related as well as suicide and alcohol-related, were in lower-income counties. So this paper made a huge splash. I think it was partially because he also concurrently won the Nobel Prize, but it was also positioned pretty well in the cultural narrative. So if you remember back to, I guess, September/October 2015, this fit in very well with the idea that Trump was becoming popular.
Luigi: Besides the Deaton and Case paper, there is another interesting paper that pushes the idea that this is an economic reason, your sort of demand-driven hypothesis, and this is a paper by two economists, Pierce and Schott, that looked at the so-called China Shock, i.e. the impact that the different treaty that we introduced with China in 1999 led to a massive increase in imports, especially in manufacturing, and this increase in imports had very negative effects in areas that were big in manufacturing, most importantly the Midwest. And so in these studies it’s difficult to establish causality, but there is certainly a strong correlation between the areas that got hit the hardest and the areas that suffered the most in terms of opioids overconsumption.
Kate: So the demand story makes a lot of sense. People were losing jobs. People were unhappy. They became addicted to drugs. What is tough about the supply story is that it’s more complex. There’s many different links in this chain. So starting with the people who were prescribed opioids themselves, there’s a question of exactly how addictive they are. And there’s been a lot of contrary evidence over the past few decades about whether opioids are indeed addictive.
And then you go up a level to doctors. What role did doctors play in their prescribing habits and who they were prescribing these drugs to? Then you go up a level to pharmaceutical companies. Were they marketing too aggressively? What tactics could they have used to influence the opioid market? And then finally, you get to the level of regulators. And there’s a question of whether there was jiggery-pokery going on, on the part of the regulators as well.
And so to really understand this whole story you have to understand the fully vertically integrated spectrum of the opioid production chain.
Luigi: Absolutely. And what might seem obvious to many people is to what extent the habit prescription of doctors have an impact on how many people get addicted to the opioids. We know that doctors have different preferences to how they are prescribing different medicines and this preference might be driven by personal preferences, different beliefs, or by how much the representative of the pharmaceutical industry is pushing hard on those doctors. Suppose that I am a patient of a doctor who is more prone to prescribe opioids. Do I get more likely addicted to opioids if I have a doctor who’s more inclined to prescribe opioids?
Kate: What’s tough to answer about this question is that it seems so obvious that, if you’ve got a doctor who’s pushing opioids on you, you’re probably going to be more likely to become addicted, but correlation doesn’t prove causation. Just based on the rough statistics, it’s hard for us to get a sense of whether there’s a causal relationship here.
Luigi: So to answer these questions, two researchers used data from Denmark. Why Danish data? Because in Denmark it is very easy to trace everybody through their social security number, and access to this very confidential data by researchers is very easy. So you can trace people very well. And what they do is they look at what happens to people who move from one area to another. And so they move from an area where the doctor does not prescribe very much opioids to an area where the doctor tends to prescribe opioids. And they see when you move, also your tendency to become addicted increases. Increases in a significant way. And that allows us to separate the story that “Oh, doctors prescribe more opioids in places where people are more desperate, and that’s the reason why you see the correlation.” No, this is like the fact that you have a doctor who prescribes opioids caused you to become more addicted, at least in probability terms.
Kate: So not quite as what we would call well identified, or clearly causal, but also in the same vein are a few studies that look at changes in regulation in the United States. There are some that look at states where medical marijuana is more available under the idea that marijuana can be a substitute in some sense for opioids. And there’s evidence that states where marijuana can be more easily obtained, opioid use was less. Another set of studies look at Medicare part D and the role that the set of laws had to do in opioid prescription. They find similar results: that there is a causal relationship between the prescriber and addiction.
Luigi: Now why does the FDA allow the use of opioids when we know that opium is a drug, we know it is highly addictive, and you cannot buy opium in the pharmacy because we know it is very dangerous? So why do we have a legitimate drug and an FDA-approved drug that contains large doses of opioids?
Kate: So even though we may know that a chemical is addictive, it’s much harder to tell whether its derivatives are also addictive and have the same sorts of properties. And to this point medical research is incredibly important because for people who, like the pharmaceutical lobby, who are looking for an answer that says it’s OK to sell this drug, even the least scientific of papers can lend legitimacy to this argument.
So to give you a good example of this: There was this 1980 letter. It was just a letter to the editor of a journal, and it only had five sentences in it, where basically a doctor looked through some old patients who had been in a hospital, he looked at those who had been prescribed opioids and those who hadn’t, and he found very little statistical evidence in terms of just like a pure one correlation that opioids were related to later addiction.
And this letter was not only like five sentences long, but also it was never intended to be taken seriously as a piece of clear scientific research. It was just a note. And yet this letter has been cited—now if you look it up on Google scholar—over 1,000 times. And it was heavily cited by the pharmaceutical lobby when they were trying to push the case that Oxycontin was a legitimate non-addictive drug. And it was also part of the training seminars for Purdue Pharmaceutical when they were training their sales representatives and teaching them that Oxycontin in fact was not addictive.
Luigi: Because here there is an important trade-off in the sense that many things that we use daily can be dangerous. Knives can be dangerous. But we’re not forbidding people from using knives even if sometimes they can be overused or used in the wrong way. And the question is what do we do with these drugs. And there is a discussion, a bigger discussion, of to what extent you want to legalize various type of drugs. I’m certainly in favor of legalizing marijuana. For more addictive drugs like opium and cocaine, that’s a different story.
But let’s assume, at least for this episode, that we want to limit the sale of these drugs in some way. Then there is the trade-off to say what are the benefits, and what are the costs? And the benefits are that these drugs are certainly useful in reducing pain. So if you have surgery or major back pain, the use of an opioid can be useful. So the question we’re trying to figure out is, in deciding between the benefit and the cost, did the regulators look at the public interest, or were they overly affected by the industry that of course stood to profit handsomely from the drug?
Kate: You may have read an article in the New Yorker by Patrick Radden Keefe about the Sackler family, who was the family behind Purdue Pharmaceuticals, the company that created and distributed and spread Oxycontin as well as a few other opioid-based painkillers. This article is just completely shocking in terms of how powerful and interconnected this single family was with every part of the opioid regulation industry, from the doctors who were prescribing this medicine to the regulators in the FDA who were supposed to oversee it.
And they would hold these training conferences essentially for doctors that were at resorts and spas, and they would fly doctors in free of cost. If this company is treating you in this really swanky way of course you’re going to be in favor of the drug that they’re promoting, especially if they’re citing research that says that that drug isn’t dangerous.
Luigi: So the good news is that Purdue Pharmaceuticals has recently announced that they are going to cut down on these aggressive marketing practices. The bad news is, the fact that they are cutting it down suggests that probably before they were being excessive.
Kate: Obviously it’s a step in the right direction. But it seems to me like putting lipstick on a pig. I know that that’s not perfect. But it’s just sort of like smoothing over something that’s already terrible.
Luigi: Yeah, but to be fair, I think that the problem is just bigger than Purdue Pharmaceuticals. It is a system that does not seem to be good at selecting things in the interest of consumers.
Kate: Yeah, and to be fair, the FDA is in charge of monitoring advertising as well as marketing of pharmaceutical products. But there’s evidence that at least historically the staff members who are in this department for overseeing this were drastically undermanned. So like in 2002, for example, there were 39 FDA staff members who were supposed to be reviewing 34,000 pieces of promotion. And there’s no way that they could have fully monitored this whole marketing strategy on the part of Purdue as well as other opioid manufacturers.
Luigi: But the producer of the drug is not the only one responsible for the situation because a lot has to do also with distribution, in the sense that the beneficiary of the drugs are not only the producers but all the channels that distribute the drugs to the individual doctors and individual pharmacies. And they in principle have a responsibility of monitoring the excessive use of the drug. The FDA knows that this drug is potentially very dangerous and as a result they require the distributors to keep track of abnormal spikes in the prescriptions. Because sometimes it’s not just a doctor influenced by the industry that prescribes too many drugs, it is somebody that really becomes almost like a drug dealer by distributing massively these opioids.
Kate: So this is all completely mind-blowing, that the pharmaceutical lobby and these pharmaceutical companies as well as distributors have such an iron grip on regulators as well as just the way that these drugs are marketed. But in doing research for this episode there was this missing link, going back to the Case and Deaton paper, about why is it that it was white people in particular, non-Hispanic white people, who were affected by this opioid crisis disproportionately relative to other races.
The demand story makes sense from that perspective, but the supply story doesn’t. Because imagine that you’re a pharmaceutical company, you would want to sell your drug to as many people as possible. But I was confused about what the differential in racial exposure to these opioid-related deaths could be. And so I looked at some scientific articles from the 2000s about this. And doctors were much more likely to prescribe opioid pain relievers to white people in any part of the prescription process, whether it was people who were receiving the pills at home or post-operative therapy, whether it was people—like, conditional on the same level of pain that people were reporting, doctors systematically were more likely to prescribe opioids to white people. And in the 2000s this was seen as a huge problem. There were a bunch of scientific articles trying to talk about how we could remove this problem of racial discrimination.
Luigi: So Kate, is this really racist or racial disparities? Because my understanding, but I’ve not read as much as you did, but my understanding is part of it is due to access to, for example, medical insurance or Medicare. You’re more likely to be prescribed if you have a richer insurance, and white people tend to have better access to medical insurance.
Kate: I think that there’s no doubt that that is true. And that’s part of the story. But just in terms of the patient-doctor relationship, another thing that really startled me about findings from the mid-’90s were that doctors were more likely to prescribe prescription opiate painkillers to patients with high-status occupations as well as patients with whom they had a close relationship. And so obviously there’s some element of trust here, but it certainly seems to be the case that doctors had in the back of their minds this idea that there was an addictive element to opioid prescription, and so therefore they were only prescribing it to people that they thought were trustworthy. And those tended to be white people.
Luigi: So in this sense this is another piece of evidence that the epidemic is actually dominated by supply. Because if it’s really economic despair driving it, there’s plenty of economic despair among black [people]. It is not a privilege or restricted to the white people who experience economic decline because black [people] experience economic decline as well. But the attitude of doctors that is different across races indicates that it is actually the supply. The more easy access you have to Oxycontin, the more addicted you get, regardless of your economic conditions.
Kate: So while you’re right that minorities, particularly black and Hispanic minorities, at the same time that this opioid epidemic was increasing, they were experiencing greater economic decline relative to their white counterparts—and so you could argue that that’s one reason that it doesn’t make sense that we would see a rise in deaths amongst middle-age whites only—but I think that the argument that many made was that it wasn’t the absolute decline in economic circumstances that mattered but it was the decline relative to the opportunities that they thought that they would have when they were younger. So it’s the decline relative to the expectation that previous generations set for them. And that was the problem.
To separate out this demand-side despair argument from the supply-side manipulation-by-pharmaceutical-industries argument, there’s a paper by Christopher Ruhm, who’s a professor at UVA, and he tries to disentangle these channels. Most of the economic variables, the economic reasons for the Case and Deaton argument, they become less significant. Even to the point where they become insignificant. So that’s a way of saying that if you include more data, then the economic reason for the increase in opioid-related deaths amongst middle-age whites, it starts to go away. And so it seems like actually the supply story is dominant.
Luigi: So to be fair to our listeners, I don’t think that there is a definite answer in this line of research, which is very recent. But the other thing we have to be careful of is we are in the process of legalizing marijuana. I think that that is great because marijuana is not as addictive, in fact is probably better than many of the drugs that are in circulation. However, at the moment this legalization happens without strong lobbying pressure by the marijuana industry, because there is not a marijuana industry, and the little that exists now is very fragmented. And in fact one could argue that the delay in the legalization of marijuana is driven by the pharmaceutical industry, who has alternative drugs more expensive and less effective, but they can be prescribed by Medicare and Medicaid.
But in the future this might change. So as we are legalizing the marijuana business we should think about do we want to put any restriction on the ability of marijuana producers to lobby the FDA or to market marijuana in the public domain and so on and so forth. I think that that ... or lobby with doctors to use more marijuana. I think that all these questions need to be addressed, and the sooner the better.
Kate: Yeah. Look, I’m in favor of legalizing marijuana as well. And by that I mean at the federal level. But I will admit that I think people can sometimes turn a blind eye to potential issues with marijuana. There are studies that establish maybe a causal link between marijuana and cognitive dysfunction, or at least cognitive impairment. And so marijuana is not a perfectly safe cure-all, and to the extent that it may be legalized in the future, we should be careful about the possibility of a powerful and influential marijuana lobby that distorts people’s perceptions of how safe this drug is, or at least perceptions of what the cost may be.
Luigi: So, Kate, suppose you were appointed the drug czar, what would you do to reduce this problem?
Kate: My ultimate objective would be for prescription opioids to still exist in very, very small quantities only available to people with extreme pain, for whom the pain is so bad that it outweighs the near certainty that they’ll become addicted to opioids. For example, people with terminal illnesses. How to do that? You have to think about, if you’re the drug czar and you want to cut back on the availability of opioid pills, you’re going to sit down at the table with people from the FDA, people from the DEA. You’re also going to get representatives from like the American Cancer Society. You’re going to get representatives from the pharmaceutical industry. And you have to have the manpower and the machinery to be able to have authority at these meetings.
And so again I know that it may be sort of a cop out of an answer, but if I were in charge I would just say like, “I want to hire a bunch of people. I want to hire a bunch of smart people. I want to make sure that everyone who’s working for me on the regulatory side is extremely well-informed. And I want to make sure that, when we go to the meetings on how to regulate this, we can face the lobbyists and the people who are on the side of the opioid industry, and make them back down.” I don’t know. Maybe that’s too optimistic or too rosy a view. What do you think we should do?
Luigi: If you want to put on our economic hats, I think that the way to solve the problem is first figure it out, how much people are willing to pay to have their pain reduced through Oxycontin versus other drugs. So you figure out what is the best alternative available, including, by the way, marijuana. And ask them, the ones who are in extreme pain, how much are you willing to pay for that difference. And then you compare to the cost in terms of lives lost due to addiction, not to mention all the cost in terms of less employability and reduced labor supply and so on and so forth.
And I’ve not done this calculation of course, but my guess is the second term would way overweight the first. And so any reasonable policy will restrict massively the availability and the prescribability of Oxycontin, and by the way, will also push for legalization and adoption of marijuana that seems to be a very cheap alternative and relatively safe alternative in this dimension.
Kate: Even though this wouldn’t cut to the heart of people’s preferences, and the costs and benefits of changing the supply of Oxycontin and other opioids, I would love to see an economic study that had very detailed data of all of the distributors in this industry, of all the pharmaceutical companies that create opioid-related pills, to know exactly how much they were spending in promotion and advertising, to know what sorts of promotion and advertising they were engaging in. And then to see the relationship between them, doctors’ prescription practices, and then people getting addicted to these drugs.
Now that data I’m sure is not available. I’m sure that the pharmaceutical companies would protect it very carefully. But if only there were an economist who did this sort of research into lobbying and special interests. I mean, gosh, who’s a person who could do that, Luigi?
Luigi: But, Kate, you have a brilliant idea. Why don’t we challenge the industry? If you have nothing to hide, show me the data.
Kate: Show us. Show us the data.
Luigi: Yes. Exactly.
Kate: I’m sure we’re going to have a bunch of pharmaceutical companies knocking at our doors.