Should oxycontin be legal? Here’s what the back of my envelope says:
In the U.S., there are about 50 million prescriptions a year for oxycontin, most of them legitimate and for the purpose of alleviating severe pain. I’m going to take a stab in the dark and guess that the average prescription is for a two-week supply.
There are also (at least if you believe what’s on the Internet) about 20,000 deaths a year in the U.S. related to oxycontin abuse. If we value a life at $10,000,000 (which is a standard estimate based on observed willingness-to-pay for life-preserving safety measures), that’s a cost of 200 billion dollars a year, or $4000 per prescription.
If those were all the costs and benefits, the conclusion would be that oxycontin should be legal if (and only if) the average American is willing to pay $4000 to avoid two weeks of severe pain. I’m guessing that might be true in some cases (particularly when the pain is excruciating) but not on average. So by that (incomplete) reckoning, oxycontin should either be off the market entirely or regulated in some entirely new way that will dramatically reduce those overdose deaths.
But of course what this overlooks on the benefit side is all the “abusers” whose lives have been enriched by oxycontin. This includes the vast majority who use and live to tell the tale, and also some of the OD’ers, for whom a few years of oxycontin highs might well have been preferable to a longer lifetime with no highs at all. Relatedly, what this overlooks on the cost side is that the average “abuser” is likely to value his life at considerably less than the typical $10 million — as evidenced by the fact that he’s electing to take these risks in the first place. Also relatedly, it overlooks the likelihood that many of those who overdose on oxycontin would, in its absence, be killing themselves some other way.
If the back of your envelope is larger than mine and you make those corrections, I’m reasonably confident that your bottom line will come out pro-oxycontin. (Please share that bottom line!) I am however, mildly surprised (and — both as a blogger who prefers slam-dunk arguments and as a libertarian who prefers to come down on the side of freedom — mildly disappointed) that the first quick-and-dirty calculation comes out the other way.
Let’s distinguish between two different questions: (1) “Would the world be better if oxycontin were magically unavailable?” and (2) “Would the world be better if oxycontin were made illegal?”
So far I think you’ve mainly addressed the first. But the second presupposes something about government behavior–for example, its likely efficacy given its track record and incentives and the long-term effects of granting it additional powers of regulation. I’m not sure about the *magnitude* of the necessary adjustments to the envelope calculation of the benefit-cost balance of a ban, but I’m pretty sure their *sign* has to be negative.
Keith: Yes, I accept your helpful distinction. Thank you.
I find these sorts of questions much easier as a comparison than as an absolute. I think most prescriptions tend to be 30 days or less. Assuming 12 prescriptions per year, that’s an annual death risk of of about 1 in 208, and 20,000 total deaths.
Consider smoking, A quick Google (we’re trusting the internet, this was first result so at least I didn’t selectively sanple) turned up https://www.bmj.com/content/suppl/2003/09/25/327.7417.694.DC1 which lists half a pack/day a day smoking as 1 in 200. The CDC among other downsides lists 41,000 annual US deaths from secondhand smoke! https://www.cdce.gov/tobacco/data_statistics/fact_sheets/fast_facts/index.htm Twice as many deaths and its collateral damage.
Looks like smoking is much worse with about the same annual death rate and twice as many deaths just as collateral damage. A similar point could likely be made about alcohol. I know this doesn’t address the cost benefit of oxy, but it points out that there are readily available and likely worse substances that should probably be addressed first.
I meant to add to previous, “if at all.”
The opioid addiction problem is hitting the UK.
Purdue basically lied about the effectiveness of the drug. They claimed it worked for 12 hours and would therefore be less addictive than other drugs. This was based on one study and at the time they had other data indicating this might not be the case. There was very little if any information on how addictive it was compared to other drugs. They aggressively marketed this to doctors, insurance companies were happy to pay the relatively low cost and doctors prescribed with apparently very little thought to cost/benefit analysis taking into account addiction.
From a SAHSA report “Approximately 11.5 million adults misused prescription pain relievers at least once in the past year, representing 4.7 percent of all adults or 12.5 percent of adults who used pain relievers in the past year (Figure 1).” 67% of these mis-uses were to relieve pain. They were presumably taking more than prescribed because it was not effective, or obtaining it illegally. This is likely to exacerbate addiction because doctors may be unaware of the levels actually taken. 12% of mis-use took it to feel good or get high.
“Approximately 91.8 million adults aged 18 or older were past year users of prescription pain relievers in 2015, representing more than one-third (37.8 percent) of the adult population.” That sounds a staggeringly large number to me, although it includes all prescription painkillers, most if not all are opiates I think.
Report here: https://www.samhsa.gov/data/sites/default/files/report_3210/ShortReport-3210.html
Whilst many of these people will not suffer from addiction, given that many over-use and the prescription limits themselves are based on bad science some sort of crisis seems almost inevitable. The imprecise use of “abuse” and “misuse” is also a problem. The above report does not differentiate. Generally, mis-use is using outside the prescription and abuse is considered non-medcal or recreational use. With these drugs I think the line will be blurred, as the motivations for mis-use are likely to be mixed. “Knee is playing up a bit – I’ll take another. Honestly, it is not because I feel shit without it, it is just for the pain.”
SL says “Relatedly, what this overlooks on the cost side is that the average “abuser” is likely to value his life at considerably less than the typical $10 million — as evidenced by the fact that he’s electing to take these risks in the first place.”
Given the above data, it seems at least possible than many abusers will have started as mis-users. They were probably prescribed the drug for some pain, had to take it more frequently or in larger doses to be effective and were reassured it was not as addictive as other drugs.
We could look at the abuse in two groups. 1) Those that started through prescription and 2) those that initially obtained the drug illegally for recreational purposes. I don’t know if this data is available.
It seems possible that those in group 2 may have tried another drug if oxycontin had not been available. However, it is unlikely that “they elected to take the risk in the first place.” I am sure the medical reassurances and widespread use in the general population was reassuring to many, who were therefore not aware of the risk they were taking. After all, whilst heroin is well know to be addictive so many people will avoid even trying it. Oxycontin was widely touted as being not very addictive, even by the doctors. “If it is OK for Mum, Dad and Granny, I am sure I will be fine” many young people may have thought.
I don’t see the “benefits” side you present as being all that large. Millions of people have slid into addiction with very little awareness of the risk.
As for the many who’s life was enriched by Oxycontin, the evidence there is weak too. San Francisco public health clinics stopped prescribing in 2005.
“What I had come to see was the lack of evidence that it was any better than morphine,” Dr. Mitchell Katz, then head of the San Francisco public health department, said in an interview.”
The cat is now out of the bag and cannot be put back. With hindsight it seems very likely that oxycontin has resulted in more harm than good. That is a different question than should we ban it now.
Similar to Keith’s argument, you have to look at government’s current role. True the average abuser values life at less than that $10 million but we all subsidize his risky behavior (higher insurance premiums if insured, more direct with Obamacare and a certainty with Medicare For All, etc.).
In the extreme, if no one dies from ODing but it costs a fortune to save them from dying, is it worth it?
I think that we should further attribute some of the overdoses to illegality. Yes, use and potential overdoses will be more prevalent if freely available, but “recreation” use could be better guided and perhaps more varied strengths available. Some of the people who misuse and die now would no longer die.
I think we need to be very careful here about our underlying assumptions. The number of deaths (20k) I assume is your estimate for the number of deaths where a prescription opioid (not just name brand oxycontin) was detected. However, there are often multiple opioids and other drugs detected when an OD occurs, so I’m not sure why you would put all the costs on one drug. The number of deaths from single prescription opioid use is a fraction of that.
I highly suggest checking out this blog to refine some of the estimates:
http://grokinfullness.blogspot.com/2018/12/drug-poisonings-update-for-2017.html (there are numerous articles at this site with great info, I highly recommend reading them all)
I think it’s overly simplistic to count the abusers’ “highs” purely as benefits.
For one thing, they usually have friends and family who remember the person pre-addiction, and while the person themselves may be happy when they’re high, it’s depressing for their friends and family to watch them disappear into this haze where they’re not themselves any more.
A libertarian might argue that it’s not relevant whether your behavior makes other people feel bad about what you’re doing to yourself, even your family and best friends. But when people talk about whether they think drugs or alcohol have harmful effects, rightly or wrongly I’m sure they’re including how they felt about seeing their family members become zombies when they were drunk or high.
@Bennett, #9,
Personal relationships tend to be reciprocal. As we value of the company of others, they will tend to value our company as well. If the opioid user accepts the opportunity cost of foregoing time with a friend or family member in order to get high, I have to wonder how much they in turn valued the company of the opioid user. That is not to belittle the experience of those to whom that happens, but I expect the aggregate social cost of the problem you describe is not a large one.
On the numbers, I don’t trust the Internet, but I do somewhat trust the CDC, and here are their numbers: https://www.cdc.gov/drugoverdose/data/index.html
According to them, 46 people die every day from prescription drug overdosing, which translates to 16,800 a year. That’s all prescription opioids, not just Oxycontin.
One problem I have with the analysis is that overdose is a broad term that encapsulates a number of situations, not all of which are solely, or even primarily, due to the pain killer used. It’s usually a combination of drugs consumed, and often some of those are illicit. My understanding is that fentanyl is responsible for a large share of the rise in opioid related deaths. The CDC says: “These reports indicate that increases in synthetic opioid-involved deaths are being driven by increases in fentanyl-involved overdose deaths, and the source of the fentanyl is more likely to be illicitly manufactured than pharmaceutical.”
My back of the envelope calculations using some more specific numbers:
According to National Institute of Drug Abuse, there were 17,029 overdose deaths from prescription opiates in 2017. (JAMA Psychiatry reported in 2016 there were 19,413 deaths from synthetic opiates, mostly fentanyl, plus another 15,000 or so from heroin. All three of these categories are included in the rising number of ‘opiate-related deaths’. Note that about 80% of these overdoses also involved alcohol or other drugs.)
In 2009, all oxycodone prescriptions accounted for about 20% of all opiate prescriptions. (The number that year was about 45 million, which corresponds well with the 50 million cited by SL) So total opiate prescriptions are about 250 million. Using the same life value of $10 million, total cost is $170 billion a year, but less than $700 per prescription of any opiate. Certainly lower than $4,000, but perhaps still high on ‘average’.
The issue I see is that this ‘average’ is a mean, whereas the distributions of these data are highly skewed. I haven’t seen published medians, but using my experience as a pain physician prescribing opiates, I have observed the following:
The majority of individuals who fill an opiate prescription receive only one or two that year, usually for a small dose, i.e., the median individual has a smaller number of prescriptions for a lower dose than the mean individual.
The risk of overdose death per individual goes up with the number of prescriptions per year, and the size of the opiate dose, i.e., the median overdose death involves people who receive more prescriptions, and higher dose opiates, than the mean number of prescriptions and strength. This distribution is highly skewed. High risk individuals receive 12 or more prescriptions a year, and their prescriptions are for opiates that are 10 times stronger or more.
The conclusion is that the median American has both less risk per prescription, and fewer prescriptions per year than the mean. Because of the skewed distributions, the median American is likely to view the cost to people like themselves as much less than $170 billion. And on an annual basis they have much less exposure to a risky prescription. My estimate is that these factors drive the perceived cost to the median American to less than $100 per year.
Further thoughts on opiates:
The above estimate of $100 perceived cost per year for the median American is for ALL opiates. Oxycodone is only one of several prescription opiates (other Schedule 2 opiates include hydrocodone, morphine, hydromorphone, oxymorphone, and some others). Oxycodone _is_ probably involved in a higher proportion of overdose deaths than it’s 20% percentage of prescriptions, but not because of Oxycontin.
Oxycontin is a brand name for one of the extended-release (ER) formulations of the pharmaceutical oxycodone. It was the first, but there are several other brands, as well as generic versions on the market. ER versions of oxycodone accounted for about 1/6 of all oxycodone prescriptions, with the trend toward a smaller fraction.
The risk of overdose, however, is not increased by ER formulation (and recent ER versions are actually safer than immediate-release formulations). The risk of overdose is related to the overall strength of the dose, i.e., how much oxycodone is contained in one pill. ER formulation risk was conflated with dose-dependent risk because Oxycontin was the first formulation to include doses as high as 80 mg per tablet.
An estimate of the abuse potential of drugs can be made by looking at the street prices. Older formulations of Oxycontin are about $1 a milligram (i.e., $40 for a 40 mg tablet), while new formulations, which are harder to crush, are down to about $0.50 a milligram. Immediate-release oxycodone is still around $1 a milligram.
By pricing measures, formulations of hydromorphone (brand name Dilaudid) are probably the most desired by abusers, since prices can range up to $80 a pill, versus all forms of oxycodone, which appear to top out around $40 a pill.
“I don’t trust the Internet, but I do somewhat trust the CDC” Yet your figures are obtained via the internet!
I suspect what you intend is that you don’t trust bloggers – quite rightly. Nevertheless there are reliable sources on the internet.
“That is not to belittle the experience of those to whom that happens, but I expect the aggregate social cost of the problem you describe is not a large one.”
I think you are belittling the experience of those to whom it happens. If your child, brother, sister or parent dies I don’t think the pain is reduced that much because they apparently did not care enough about you because of their addiction.
@Harold, #13:
“Yet your figures are obtained via the internet!” Just because I don’t trust Donald Trump does not mean that he will never say something that is coherent and true.
“I think you are belittling the experience of those to whom it happens”
It is possible to lament the huge personal losses in individual cases, while pointing out a small total cost across a group. I speak from personal experience. I have known many recreational drug users. It is generally true that those that decided getting high was more interesting than hanging out with others were also the ones whose company were missed less by the group. Also, those whose company *were* missed were also ones where friends were more likely to intervene, perhaps preventing something tragic happening.
“Just because I don’t trust Donald Trump does not mean that he will never say something that is coherent and true.”
Yes, it does still remain a possibility.
On the life value aspect. SL says “Relatedly, what this overlooks on the cost side is that the average “abuser” is likely to value his life at considerably less than the typical $10 million”
This is probably true after they have become addicted, certainly by the willingness-to-pay for life-preserving safety measures. But do we have any information before they were addicted? After all, that is one of the effects of the drug. If the addiction results in a drop in the value of their life from $10M to $1M, is that not a cost?
The main benefit of Oxycontin in particular was that the users would need to take it only twice a day, and that claim is very suspect. There is also evidence that the 12 hour cycle actually adds to addictive nature. The benefits over other, much cheaper alternatives is not so much the removal of pain, but greater convenience. I think Oxycontin itself offered poor value for money. We could have had almost the same results for much less with morphine.
In 2001, Portugal decriminalized all drugs.
It’s practically a lab experiment.
Why is no one interested in learning?
I’d like to see the back of envelope calculations on marriage, kids, religion and pets. I’d pay a lot to get rid of them all.
A quick search says after Portugal decriminalized all drugs, their overdose incidents decreased to about 3 per every million people. Compared to the rest of the EU of 17.3 per million.
A few comments.
Trent McBride links to a post of mine with some figures and commentary on overdose deaths. In 2017, there were 12,308 accidental overdose deaths involving “Other Opioids”, a class which includes Oxycontin and many other commonly prescribed opioids. As some commenters point out above, strictly Oxy-related deaths are a fraction of this number. It’s not in my post, but the number of total deaths involving “other opioids” in 2017 is 14,717. Many of these are suicides, “undetermined intent”, or aren’t even overdose deaths. I think the number that’s relevant for this post is *accidental* deaths, or 12,308. Otherwise we’re speculating that “but for Oxycontin, those people would not have committed suicide.” (Anyone can look up numbers on the CDC website using their Wonder database.)
http://grokinfullness.blogspot.com/2018/12/drug-poisonings-update-for-2017.html
Many of these deaths are probably more appropriately labeled “heroin overdoses”, not Oxy or prescription opioid overdoses. Most drug poisoning deaths aren’t actually an overdose of a single substance, but are rather due to a combination of drugs, and the death certificate will list all the substances involved. If you exclude “other opioid” deaths that also involve heroin or synthetic narcotics (which mostly means street “heroin” that is actually fentanyl), it’s down to 8,691. And even with this subset, most of these are more properly thought of as “opioids interacting with benzodiazepines” or “opioids interacting with alcohol.” These aren’t exactly suicides, but they are mostly examples of people deliberately engaging in very reckless behavior. It’s not like everyone prescribed opioids has a 10,000-chamber revolver pointed at them, as if they’re being subject to a risk that is outside their control. Most patients either take their opioids as prescribed or actually stop taking them before they run out, but a small subset very deliberately engage in risk-taking.
The 20,000 is clearly too high, but maybe the post is salvageable by generalizing it from Oxy to other prescription opioids and reducing the number of deaths. The bigger problem is treating an Oxy misuse/overdose like it’s a stochastic risk, rather than a deliberate choice. Landburg nails it when he says:
“Relatedly, what this overlooks on the cost side is that the average “abuser” is likely to value his life at considerably less than the typical $10 million — as evidenced by the fact that he’s electing to take these risks in the first place.”
Or, forgetting all that, suppose I take $4,000 to be the right figure for “cost per prescription.” (Or suppose I come up with a similarly high figure some other way.) This number is based on the cost paid *by the user* in terms of risk of death. If I were in charge of drug policy, I wouldn’t look at that bottom line and say, “The cost is high, this must be stopped!” I’d instead react by saying, “This is going to be very hard to deter, considering the very high costs people are willing to pay.” Other commenters above made a similar point about prohibition being hard. Prohibition would have to do something very severe to deter people who willingly pay such a high cost. Recent attempts to crack down on prescription opioids have led recreational users to buy heroin and fentanyl, which are far deadlier. You can see the results in the charts in my post. (2010, when “abuse-resistant Oxy” hit the market, is the year when heroin deaths started to rise dramatically.)
“Most patients either take their opioids as prescribed or actually stop taking them before they run out, but a small subset very deliberately engage in risk-taking.”
The study I quoted supports the fact that most people use their meduication correctly, but contradicts the “small sub-set.” In 2015, 12.5% of pain relief users misused. That is 11.5 million people. This is a big sub-set. That was 4 years ago, and I don’t think it has got better since then. Obviously not all these go on to have problems, but we have to acknowledge that misuse, which can lead to addiction, is a big problem.
If we take an individualist view, then addicts are simply exercising their free choice to engage in risky behaviour. However, we know that if we make opiates freely available then people are going to get addicted.
So my earlier question, if a person values their life at $10M, then gets addicted and then values their life at $5, have they made a loss of about $10M?
I’m late to the party, but doesn’t the answer to Steve’s query depend on the overall amount of harm caused by oxycontin, compared to the overall benefits produced?
Oxycontin should be more “legal” i.e. the marketing(education) to physicians and patients should be more regulated. Supplier induced demand may account for a large portion of the abuses of this drug. There are many lawsuits against the manufacturer of this drug(some already settled) alleging unethical (and highly successful) practices to increase sales.