In the first ever episode of In Plain English, presenter Jamie Moffa and guests Brendan Ziebarth and Alex Marshall discuss the paper “Impact of co-administration of oxycodone and smoked cannabis on analgesia and abuse liability” by Ziva Cooper, et. al. They go over the promising possibility of using cannabis to lower the dose of oxycodone needed for pain relief, the tests the authors did to measure abuse likelihood, the importance of looking at funding sources, and more.

Article PDF

Episode Transcript

[Intro music]

Jamie Moffa: Hello, everyone. Welcome to In Plain English, a podcast where we discuss scientific research in terms that are accessible to everyone, not just the experts. My name is Jamie Moffa. I am a PhD student studying neuroscience. Specifically, I’m researching how the brain responds to pain and opioid medications. I’m also your host for this podcast and will be presenting today’s scientific paper. Since this is the first episode, I’m going to take some time to go over the format, where you can find more information, and how you can continue the conversation once the episode is over.

Every episode of In Plain English will feature an expert in a particular field of science, such as genetics or anthropology. The expert will then present a paper in their field and the podcast guests, people without much knowledge of the subject, ask questions and discuss the research. Every episode will conclude with a broader discussion of how the paper’s findings impact society as a whole. You can download the paper for each episode at inplainenglishpod.org by clicking on the episodes link in the main menu. We believe in open access science for all, so the papers we choose will always be free for you to download. We will also have transcripts posted on our website for each episode. If you have a question or comment about a previous article, you can submit it under the Continue the Conversation tab. In future episodes, we will begin by reading and responding to some of the questions and comments that you send in. If you are interested in being a guest or presenter for a future episode, you can click on the Become a Guest tab on the website. You can also reach out to us on Facebook or Twitter at plainenglishsi, that’s P-L-A-I-N-E-N-G-L-I-S-H-S-C-I.

With that out of the way, on to today’s paper.

JM: Today, I will be presenting a paper on the pain-relieving effects of marijuana and the potential for medical marijuana to reduce the dose of opioid pain medications like oxycodone needed to manage pain. I am joined by my wonderful guests, Alex Marshall and Brendan Ziebarth. Alex, Brendan, welcome to the show. Would you each like to briefly introduce yourselves?

Brendan Ziebarth: Sure. Thank you so much for having me, Jamie. My name is Brendan Ziebarth and I am currently a nursing student. I’ve had several other careers before this, but I’m excited to be here and to discuss this paper with you.

Alex Marshall: And I’m Alex. Thank you for having me here. I am a genetic scientist for Yale University. I know absolutely nothing about drugs and impact, so it was very interesting to read this paper and get a little bit of an idea of it, and I’m excited to get started.

JM: Great, and I want to thank you both again for being guests on this very first episode of In Plain English. So without further ado, let’s get to the science.

So as I mentioned earlier, this paper asks the question, can smoking medical marijuana reduce the amount of opioids needed for pain relief? And given the ongoing opioid crisis in the United States, the side effects of these medications, and the potential to develop an addiction to opioids, examining whether cannabis can be used as a supplement or a substitute for opioids is critical. And as the authors mentioned at the beginning of their paper, there is anecdotal evidence, meaning evidence that comes from people’s personal experiences, that marijuana can be used as a substitute to opioids in order to manage pain. Their goal is to determine whether this is the case in a more controlled setting.

But before we get into what the researchers did and what results they found, we thought it might be useful for Alex and Brendan to give us a list of terms to know.

AM: When I was reading this through, my first question was really what are opioids? Why are we studying opioids? The basic information I had before going in here was they’re for pain and they’re addictive, and that’s about it. So can you explain a little bit more about what opioids are?

JM: So opioids, that is the general term for a class of pain medications that are derived from opium originally, but the most common ones used medically are morphine, oxycodone, and derivatives of those. Opioids can also be drugs of abuse or recreational drugs. Basically, what these drugs do, among other things, is to act on your brain and your sensation system in order to reduce the amount of pain that you feel, but they also have a variety of other effects, which is why people also use them recreationally.

AM: Thank you, Jamie. Some of the things that I had to look up, one of them was cannabinoids?

JM: Cannabinoids.

AM: Basically, it’s just the stuff that is in marijuana. There’s specific chemical names that we can and can’t go through, and I’m sure that we’ll touch on that later, but cannabinoids is the things that are in marijuana.

BZ: Active ingredient, if you will.

JM: Yes, it’s helpful to mention here the one that’s going to be the most talked about in this article, and also the one that the most people know of is THC, which is the major active ingredient in marijuana and is the one that gets you high.

BZ: So, nociceptive stimuli is going to be any kind of stimulus, any kind of action that causes a painful reaction in your body, the sensation of pain. Nociceptors are the nerve endings that detect pain in our bodies. So, nociceptive stimulus is one that causes us to feel pain.

AM: The next one that really started to trip me up was analgesia, and that is a lack of pain. An analgesia effect would be something that stops or reduces pain.

BZ: I’ll talk a bit more about anti-nociceptives. So, synergistic anti-nociceptive effects work together to reduce the amount of a drug needed, a drug that produces analgesia. The amount of it can be reduced if there are synergistic effects between analgesic drugs. The lowest effective anti-nociceptive opioid dose is the lowest active dose of a opioid that one would have to take in order to produce analgesia.

AM: So, just to clarify, just the lowest amount you need for it to be effective?

BZ: Correct, yeah.

AM: One of the other things that we’re going to be talking about is opioid sparing effects, and that runs into the lowest effective anti-nociceptive opioid dose. So, when you have an opioid sparing effect, that means it has the effect of needing less of the opioid to control the pain.

JM: So, for example, in this paper, the authors wanted to show that marijuana has an opioid sparing effect.

BZ: We also looked into abuse liability, which is the likelihood that someone will abuse a drug based on the drug’s properties.

AM: This particular study is a placebo-controlled double-blind within-subject study, and I’m going to break down those for you. So, a placebo-controlled study is a study that has a substitute instead of an actual active ingredient. So, it would be, for example, taking a sugar pill instead of an opioid or smoking something that doesn’t have THC versus something that does have THC. So, they’re taking out the controlled substance and replacing it with something that doesn’t have an effect at all, and that’s what a placebo-controlled study is.

A double-blind study is where the scientists don’t know what subjects have the real active ingredient and which subjects don’t, and the subjects don’t know if they have the real active ingredient or if they don’t. So, neither the scientists nor the people who are taking part in the study know who has what, and that makes it double-blind. It takes away a lot of the…

BZ: Takes away the bias and the confirmation bias that a scientist might be looking for the effects in a certain person if they knew that they took a drug that they expect to have a certain effect, but if neither the scientist nor the person who took the drug know if they got the placebo or the real drug, then it helps eliminate that confirmation bias. Within subject studies are studies in which all participants experience all experimental conditions. In this particular study, there are six possible combinations of drugs. The possible conditions in this study are: receiving placebo opioid and placebo marijuana, receiving marijuana and placebo opioid, receiving placebo marijuana, and the 2.5 milligram dose of opioid, receiving placebo marijuana and the five milligram dose of opioid, or receiving active marijuana with the 2.5 milligram dose of opioid, or receiving active marijuana with the five milligram dose of opioid.

In a within subject study, the results from each condition are analyzed against the same individual’s response to the other condition. So this creates a lot more statistical power and predictive ability because you’re eliminating other variables that come up by having different subjects taking different combinations of the drugs, such as the fact that each person’s body is going to respond somewhat differently to any amount of the drug. By comparing these results within the same subject, all six experimental conditions, we get a much better picture of what is actually the effect of the drug and eliminate some of that background noise.

AM: I also had to look up what a cold pressor test is, and that was one of their main formats to test pain. So a cold pressor test is when you take your hand, you stuff it into a bowl of ice water and you say, hmm, it’s starting to hurt and then you start counting and however long you can stand your hand being in that ice water after it starts hurting and then you take it out. So the two things that they’re measuring there is pain threshold and pain tolerance. So when you take your hand, you stuff it into a thing of ice water, how long until it hurts is your pain threshold. The amount of time you stick your hand in the water until you say, hmm, that hurts. Your pain tolerance is after it starts to hurt, how long can you keep it in there before you take your hand away?

JM: With those terms under our belt, let’s talk about the main experiment that the researchers performed in order to answer their question. For this study, they recruited people who regularly smoke marijuana. And this was in order to avoid any unknown effects from first time use. And all of the participants had also taken opioids at some point in the past. During the study, each participant underwent the cold pressor test, which as Alex described, measured participants’ pain threshold and their pain tolerance by having them hold their hand in ice water. So they did the cold pressor test before they took any drug, after they took capsule containing the opioid oxycodone, and after they smoked a cigarette containing a set amount of cannabis. So the participants’ pain threshold and pain tolerance after taking the opioid and after smoking marijuana were compared with their initial results in order to see whether either drug or both of them in combination helped alleviate their pain. Longer times before reporting pain and longer times before removing their hand were considered metrics of pain relief.

AM: Now, Jamie, you mentioned specifically that they did oxycodone. Is there a significance to oxycodone versus other drugs that they could have taken in terms of opioids? Did it have to do with patient history?

JM: I don’t think it had to do with any of their specific histories. It has to do with the fact that oxycodone is one of the most commonly prescribed opioids in a medical setting.

So in addition to comparing the effects of opioids and cannabis to those participants’ baseline result, the researchers also wanted to compare the effects of different amounts of opioids and cannabis on pain relief. And Brendan went over this beautifully, and as they mentioned, the participants took either 0 milligrams or a placebo of oxycodone, 2.5 milligrams, which I will refer to as low dose, or 5 milligrams, which I will refer to as high dose of oxycodone. And the marijuana was either a placebo containing no THC, which is the active ingredient, or 5.6 percent THC, which I should note is actually a very low amount of THC compared with commercially available marijuana products.

BZ: I wanted to point out that commercially available marijuana products that you can get at a medical dispensary or recreational dispensary regularly advertise THC percentages between 18 and 30 percent compared to 5.6 percent THC, which is what was used in this study.

AM: So that is fairly low, then, comparatively speaking.

BZ: It is very low. Apparently, one of the confounding factors in this study is that it’s very difficult to study controlled substances in the United States, including marijuana. It takes a lot of government regulation and oversight, and the government provides the drugs that are used in the study. So the government is providing a kind of marijuana that is grown in a government laboratory somewhere under very controlled conditions. They are using strains of marijuana that have a very low percentage of THC that are not really reflective of strains that are commonly available on the street or in recreational or medical dispensaries.

AM: So now do you think that this would affect the way that people see and take in this study?

BZ: Well, one of the things that I think they did well in the experimental design is select current cannabis users, people who regularly use cannabis. And they discuss in the demographic section that the average amount of money that their research subjects spent per week on marijuana was $153, which is a significant amount of money to spend on marijuana. And they also translated that to a aproximate number of marijuana cigarettes smoked in a week for each subject. And that came out to somewhere between one to two a day on average. So these are regular marijuana users, presumably using more concentrated THC. One of the things that I looked for in this study was to see if the low percentage THC marijuana used had a measurable or perceptible effect on people who are regular cannabis users. I as the reader expected that the regular cannabis users would report low perceptibility of the effect of the marijuana because they’re used to much higher doses and much higher percentages of THC. And the results actually seem to indicate that the subjects were able to perceive the marijuana that was active as opposed to the placebo with near 100% accuracy. I think it is pointing towards the fact that even low doses of marijuana may be effective in relieving pain, even in regular cannabis users.

AM: Well, I think that neither the active cannabis nor the low dose of oxycodone were therapeutic in any way, therapeutic meaning that it worked to relieve the pain. But only when they were together did it create an analgesic effect.

JM: I think we can use that to kind of segue into the results that they found from their main experiment. As you pointed out, so in any of the conditions, whether they were measuring pain threshold or pain tolerance, the active cannabis did not have a impact on the participants’ time to report pain or time to remove their hand from the cold water. And this could be due to the fact that it was a very low dose of cannabis. This could be due to the fact that maybe cannabis does not reduce this kind of pain or it could be due to some other factor. But that low dose of oxycodone, which also was not effective on its own when combined with cannabis, did produce a significant effect.

And I want to take a moment to talk about what that word significant means. In research, in science, the word significant doesn’t just mean dramatic or important. Scientists will utilize statistics and probability in order to determine whether their experiments really caused the changes that they observed or whether those changes happened by chance. And to better illustrate this, because I am a gaming nerd, I’m going to use an example of two six-sided dice. You have a green die and a pink die. And you believe that the green die is lucky and tends to roll sixes more often than the pink die. And you want to test this out, so you roll both of the dice. And sure enough, the green die rolls a six and the pink one rolls a three. But any six-sided die that has the numbers one through six on it will roll a six at least some of the time. So the fact that the green die here rolled a six and the pink one didn’t doesn’t really mean much. So if you roll the dice again, and once again, the green die rolls a six and the pink one rolls a five this time. Now this is a bit of a coincidence. Like, oh, the green die rolled a six twice in a row. But it’s still not incredibly uncommon for a six-sided die to roll two sixes in a row. What counts in research as quote, incredibly uncommon? How much do scientists have to roll their experimental dice, either by recruiting more subjects or doing more repetitions of their experiment in order to feel confident that the results that they got were not just due to random chance. In most fields, this level of confidence is set to 5%. So scientists are confident in saying that their results are significant, meaning likely to be caused by the things that they did and not just by chance, if there is a five in a hundred or one in 20 chance that they got it wrong.

So in this paper, they used this 5% level, meaning that based on the probability calculations that the authors did, there is a 5% chance that participants’ pain thresholds just increased randomly and that the drugs actually did not have an effect. Or put another way, for the combination of the 2.5 milligrams of oxycodone and marijuana, there is a 95% chance that those two drugs together actually did increase participants pain thresholds. And the researchers saw a similar effect on pain tolerance, which was again the amount of time that participants were able to leave their hand in the cold water before removing it, where once again, the cannabis on its own and the low dose of oxycodone on its own were not enough to increase this tolerance, but the combination was.

And when it came to the high dose of oxycodone, so the 5 milligrams of oxycodone, that on its own did increase pain threshold and did increase pain tolerance. And adding cannabis—so the cannabis plus high dose of oxycodone also increased pain threshold and pain tolerance, but not any more so than the high dose of oxycodone on its own.

AM: When I was reading, I noticed that they were using subjective ratings. And I know that this goes back to what you were just talking about and the accuracy of the testing. And I wanted to ask you what is good and what is bad about using subjective ratings?

JM: So the subjective ratings was actually a second thing that the researchers did after the participants had completed this cold pressor test. So the cold pressor test on its own and the times for the participants to report pain and remove their hand from the water are called objective pain measures, meaning that it’s sort of hard and fast, set in stone. There’s a concrete thing that you can measure and report. And the only kind of wiggle room is just in sort of people’s individual differences in how well they handle pain. So the subjective pain metrics that the researchers used after the cold pressor tests are asking the participants to report how bothersome the pain was, how painful it was, and also a number of other dimensions of pain. So that’s getting at not just how the participants reacted to pain in the moment, but how they thought about it and how they experienced it.

And it’s important to have both subjective and objective pain metrics when you’re studying pain, because oftentimes in research we can get tricked into thinking that a new drug for pain is effective because it reduces those objective measures of pain. It reduces pain as measured by something like this cold pressor test. But if we don’t bother to see how it reduces the subjective component of pain, then we could end up taking that drug to clinic and have people report that their pain doesn’t seem to be reduced very much even if by some objective metric it is. However, so there are issues though with using these subjective pain metrics, one of them being that it is in fact subjective, meaning up to each person’s interpretation of what very painful means. The painfulness and bothersomeness scale were reported on a scale of one to ten, much like your doctor would ask you. What each person defines as a one or a five or a ten on that pain scale will differ. Another issue is that since this was performed after they did the cold pressor test, there could be issues with their recall of the pain in that they’re trying to remember how it felt instead of saying how it felt in the moment, and that can affect how they respond as well.

BZ: Right, I wanted to bring that issue up specifically in regards to the McGill pain questionnaire, which the study procedure describes as a “15 item shortened computerized form used to assess the sensory and effective dimensions of the pain,” asking people to describe what in nursing we refer to as pain descriptors, such as throbbing, shooting, stabbing, and these scores were given with one being none of that quality of pain, and severe being graded a four for that quality of pain. But then these scores were added up across all 15 items to generate a sum score ranging between 15 and 60. And these questionnaires were completed immediately after the cold pressor test. So immediately after dunking your hand in the water and then taking it out, you take—you immediately after take this 15 item McGill pain questionnaire. What concerns me with that methodology is, that person is attempting to recall the quality of pain after the pain has already disappeared or mostly disappeared, and attempting to apply descriptors to it—the pain retroactively. That seems fraught with recall error and likely to invite people to sort of answer fictitiously in order to complete the requirements of the study, rather than seeming to capture meaningful data in the moment.

JM: I think that’s an important point to bring up. Yeah—

BZ: Let’s just point out that this was a six hour and 15 minute day, and they were performing multiple different questionnaires and assessments throughout those six hours and 15 minutes. So I can imagine this was a grueling day with probably more forms to fill out in one day than I may have ever done in my life. Beyond the McGill pain questionnaire, they also had subjective drug effects.

JM: So there were three of these metrics for cannabis and four for oxycodone, but the three for cannabis were strength, how much they liked it, and how high it got them. For oxycodone, they did a similar thing, but with four different metrics. So again, they used strength. They had a metric called good drug, which I think was sort of related to high, how much they liked it, so similar to cannabis again, and then whether they would take it again. Why they were asking people how good the drug was and if they wanted to take it again and if it got them high, was because they wanted to see if giving cannabis and oxycodone together increased people’s likelihood to abuse either one, where abuse is defined as using a drug more often or in higher quantities than prescribed say by a medical professional. They wanted to find out if people were likely to use these drugs for something other than pain, were they to be given them in a medical setting.

AM: I think this is a great segue into talking about abuse liability and self-administration and how to kind of navigate that difficult field in a scientific study.

JM: Let’s talk first about what they did for self-administration and then we can talk about the second part of that question. In addition to all of these other questionnaires, they had participants basically be able to pay money in order to take additional puffs from their marijuana cigarettes. The reason why they did this was to see, again sort of similar to all of these other metrics, how much people would take it in excess of what was needed for pain relief, when given the opportunity. And what they found was that people were less likely to do this with inactive cannabis, and that was regardless of whether that cannabis was administered with a placebo oxycodone or with either of the two doses. People were not very likely to smoke more inactive weed.

BZ: This comes back to a point that I was bringing up earlier about the percentage of THC and whether regular smokers would feel an effect from a lower dose of THC. I’d mentioned earlier that the regular smokers were in fact able, with near 100% accuracy, to determine which of the cigarettes was placebo and which was active marijuana.

AM: I also want to clarify, they can tell what is and what isn’t active cannabis, but the active cannabis was still sub-therapeutic. So having an analgesic effect or not having an analgesic effect is separate from being able to tell whether it is an active or an inactive cannabis.

JM: That’s a good clarification. Moving from our discussion of inactive cannabis, which they did not pay to self-administer very much, active cannabis they paid more to self-administer it and took more puffs from the active cannabis cigarettes, going to Brendan’s point that they could in fact tell that this was an active drug versus a placebo. The addition of oxycodone, while it did somewhat increase the amount that they paid to administer more puffs, this was not significant. So going back to our discussion of significance, the researchers basically couldn’t determine whether that was a real effect or whether it was simply sort of within this range of just random chance. So I think that this particular figure about the self-administration hits the closest to what they were trying to get at with abuse liability and likelihood that someone would use a drug in a higher quantity or a higher frequency than prescribed.

BZ: Can we actually pick apart figure three real quick?

JM: Oh sure. So looking at this figure, we’ll start with the figure axes. So on the horizontal axis or the x-axis we have the grouping of inactive cannabis trials, so trials in which the fake marijuana cigarette with zero THC was given either with a placebo oxycodone, with the low dose, or with the high dose of oxycodone. And then to the right of that we have the active cannabis trials where 5.6 percent THC was administered again with placebo oxycodone, low dose, or high dose. On the y-axis or the vertical axis we have puffs, which are measured in dollars, and according to the figure legend each extra puff cost one dollar. So this is measuring basically, for each dollar increment on the y-axis is another additional puff that the participants chose to take. We can see that when inactive cannabis was administered with a placebo oxycodone—so this person was getting zero drugs—people on average took about half of a puff. So some people took one additional puff and paid one dollar. Some people took zero additional puffs and paid zero dollars. When this inactive cannabis was administered with either the low dose or the high dose of oxycodone, people on average paid one dollar to take one additional puff. Statistically speaking people did not take a significant number more puffs of the inactive cannabis when it was administered with oxycodone versus not.

With the active cannabis, when it was administered with the placebo oxycodone, people took on average slightly more than one additional puff, and this was significant meaning it was likely that this was a real effect of the active cannabis when compared with the inactive cannabis plus placebo oxycodone trial which was used as the baseline.

BZ: So basically when people could tell it was real weed as opposed to fake weed they wanted on average to buy one more puff.

JM: And when that real weed was administered with either low dose or high dose of oxycodone, people on average wanted to take 1.5 more puffs, which again was significant compared to the inactive cannabis plus placebo condition, but was not significant compared to the active cannabis plus placebo oxycodone condition. People want to smoke real cannabis more than they want to smoke fake cannabis, and they can tell that it’s real cannabis or not and this is not impacted by whether or not they also had oxycodone.

BZ: So even though graphically it does appear that there is a difference between if you take either a low or high dose of oxycodone, you are also more likely to want another puff of the active cannabis, the statistics are not saying that there is a significant difference between taking the oxycodone and wanting another puff of the active cannabis.

JM: Correct.

AM: So this is really a poorly done graph, because it looks visually like there’s a significant difference when, if you look at the statistical analysis they’re saying that it is not a significant difference.

JM: Yeah so I think it’s tricky with this kind of graph and with this kind of experiment because of the low number of people who they had in their study and because of the low number of options of how many additional puffs they could take: basically zero one or two. The difference of half a puff of a marijuana cigarette looks large; when considered either statistically or realistically is not actually that large. There are a couple of things at play here. One is to make sure that you look at not just the graphical representation but what the researchers actually said they found, and two is to look at the actual real world difference in effects that are being reported. Something could even be statistically significant—

BZ: If you just glance at this graph as a casual reader you might interpret that taking oxycodone increased the likelihood that you would want to purchase more puffs of the marijuana cigarette, but that is not actually what the statistics show is the significant effect. It’s the fact that if you’ve smoked the active cannabis cigarette, you are more likely to want to purchase another puff.

JM: I—imagine yourself being a layperson science reporter and you wanted to report on this study and you see this graph, you can imagine how someone would see that and say “oh oxycodone looks like it increases people’s likelihood of using more cannabis,” and then run with that headline, which would be incorrect.

I found this paper because I heard the first author of the paper give a talk. She had said that there were so many issues with previous research about cannabis and opioids and pain relief that she was like, “all right, we’re going to design one that is actually controlled and double-blinded.”

BZ: The biggest difference of this one is also that you had—it’s placebo controlled and you had two levels of opioid administration, so you could tell if the cannabis actually reduced the effective dose of the opioid, as opposed to other research, which has just basically asked subjectively “when we combine these to do you feel pain relief?”

JM: Which is the critical finding of this paper in my opinion at least: that people had similar experiences of pain relief with the cannabis plus low dose of oxycodone as compared with the high dose of oxycodone alone, or the cannabis plus high dose of oxycodone, which shows what the authors wanted to show, which was that cannabis can have those opioid sparing effects where it makes you need less of the opioid in order to have the same amount of pain relief.

BZ: Circling back to our discussion on abuse liability and whether the methods of this paper actually determined anything useful about abuse liability. Table four as far as I understand it is where they present the data of the abuse liability of oxycodone.

JM: So they were comparing each of the six different conditions that we have talked about. Under the subjective effect they are measuring strength, how good the drug was, how much people liked it and how much they would want to take it again, and participants rated these all on a one to one hundred scale. They were comparing each of the conditions to the zero milligrams of oxycodone and zero percent THC, or placebo condition, which was sort of a baseline of nothing’s really going on here. And they wanted to see if administering opioids with cannabis would increase these ratings over just the opioid alone or no opioid at all. What they ended up finding was actually that it was a small change but a significant one from a statistical perspective that administering the low dose of oxycodone with the 5.6 percent THC cigarette did increase the participants’ rating of the drug strength, how good it was, how much they liked it, and whether they would take it again, both compared to control obviously, but also compared to the low dose of oxycodone administered on its own.

BZ: I’m understanding it differently. I’m understanding that the 2.5 milligram column, 5.6% cannabis column—so low dose oxycodone and active cannabis—is being compared against the placebo of both, and being compared against active cannabis alone, and that’s where they have significant results.

JM: Because I had assumed that you would want to compare it—if what you want to say is that adding cannabis makes you more likely to use more oxycodone, then to me what you would want to compare it to was what I said, which was comparing your low dose oxycodone with and without cannabis. But what they actually did was compared their low dose oxycodone with cannabis to cannabis without oxycodone. I don’t think in that case that that’s showing what they want it to. What they actually did was just compare everything to placebo, and then just compare what it looks like they did is they just compared everything then to the 5.6 percent THC and zero milligrams of oxycodone.

BZ: What does that practically mean? I mean you’re comparing—so when you’ve got a little bit of oxycodone and active THC, you are more likely to want to take the oxycodone again than the placebo. And then the same can be true said to be true versus cannabis alone. So if you have a low dose of oxycodone plus active THC you are more likely to find it to be a strong drug than THC alone. You are more likely to find it to be a good drug than THC alone. You are more likely to like it than THC alone and you are more likely to take it again than THC alone.

JM: What I’m saying is that I actually don’t understand why they did it that way, because if they want to show that like whether or not adding cannabis increases your liking of oxycodone, what they should do is compare the oxycodone at the same concentration with and without cannabis, and then see if that changes.

BZ: I think you’re right, like I don’t think they did the right comparison because I don’t think they’re really testing the cross effect. Okay so can we pick a part figure two now?

JM: Okay, so this is the strength, liking and high of active cannabis related to inactive cannabis.

BZ: Okay so this is like where we can actually start to see a separating out like drug effect. The regular cannabis users ranked the strength of both placebos very low, close to zero, whereas they ranked the strength of cannabis alone as somewhere between a little over 60 out of 100 to a little under 60 out of 100.

JM: Yeah.

BZ: Now when they added in oxycodone at either dose, but with the inactive cannabis, people ranked the strength of the cannabis—so this is placebo cannabis but they did have oxycodone—they ranked the strength of the cannabis as a little bit over 10 maybe 15 to maybe a little bit below 10 or almost 20. Some people did perceive a cannabis effect—a cannabis like effect from the mystery pill.

JM: However I don’t think that—I think that they compared all of these conditions to placebo and placebo and none of those reached—

BZ: But they weren’t statistically significant.

JM: Yeah.

BZ: And then when you compare these active cannabis plus oxycodone it’s all very very similar. The lines are like very close together.

JM: Yeah.

BZ: The same goes for strength, liking, and high, that basically people just could tell whether this was active cannabis or not regardless of the effect of the opioid. So it ends up that like in all three of their data analyses that are attempting to show abuse liability, they’re actually just comparing the active drug to the placebo. I don’t actually see any significant results that are showing an interaction of the drugs to create more abuse liability.

JM: Right and so I think that that—and they say this in the text—but I think that they take that as meaning that oxycodone does not increase the abuse liability of cannabis. So people are not more likely to smoke or intake more cannabis if they’re also taking oxycodone.

BZ: How—it does say though, “However, subjective ratings related to oxycodone abuse liability showed small but reliable increases after active cannabis administration.”

JM: But that’s not the comparison that they said that they did.

BZ: They didn’t run the analysis on the same doses of oxycodone with or without the cannabis.

JM: I would think that that would be the relevant comparison.

BZ: It’s basically just coming from this like pleasure shaming perspective, which is like, if you can detect an effect from the drug and it is in any way pleasurable, they’re saying that you’re more likely to abuse it and that’s basically all that they’ve said here.

JM: They actually say in the text that they specifically want to minimize that kind of an effect. The frustrating thing to me is that they’re claiming that cannabis increases the abuse liability of oxycodone, but they have not actually tested that as we have been talking about extensively. I think it’s interesting that they did these comparisons to cannabis alone and not between oxycodone conditions or within oxycodone conditions, in the context of what Brendan found about their funding sources, because it seems—

BZ: That’s exactly what just came to mind for me too! Yeah so you know I am also a very political person, and I always follow the money when I can and I do appreciate that these studies have disclaimers about the funding. There’s a statement here that the primary author and then the last author have received research funds and partial salary support from INSYS therapeutics, but I take this first sentence to mean INSYS therapeutics actually provided funds for this specific study. That’s a large financial commitment to this study, like INSYS therapeutics is invested in the results of this study. So this study came out in 2018, and INSYS therapeutics is an American specialty pharmaceutical company based in Chandler Arizona. Its main product is Subsys, which is a sublingual liquid form of the drug fentanyl which is a very powerful opioid. In 2019 INSYS filed for bankruptcy with the first author and the last author actually listed in its bankruptcy filings as creditors along with 13,000 other creditors, many of which were doctors, many of which were other seemingly institutional investors including cities. But in January of 2020, the founder of INSYS therapeutics named John Kapoor was actually sentenced in federal court for orchestrating a scheme to bribe medical practitioners to prescribe Subsys, the fentanyl based pain medication, often when medically unnecessary. This is a company that was bribing doctors to prescribe fentanyl in order to increase their sales revenue at the height of the opioid epidemic. And this is the same company that is funding this study.

AM: Seems like a weird motivation to lower the effective dose needed of opioids by using cannabis as well because wouldn’t they want to sell more opioid?

JM: I highly doubt that INSYS was directly putting pressure on the authors to find that in fact cannabis did not lower the needed dose of opioids. That would be I think too cartoonishly bad, and that does happen in scientific research sometimes. But the fact that the comparisons that these researchers did and the fact that they found that, or tried to imply, that cannabis increased the abuse liability of oxycodone—

AM: Even though they never tested that.

JM: Even though they never tested that, but that they stated that they found that, based off of a different test entirely, I think is possibly one avenue of this influence, because they’re trying to show that it could be dangerous to administer cannabis with oxycodones, and cannabis at least at the quantity tested here was not sufficient to produce pain relief, so if it’s dangerous to administer both of them together, and cannabis alone is not effective, then you might as well continue prescribing only opioids.

BZ: I think that’s really interesting potential reading of the relationship. You know another potential reading is that I don’t know if this is true but like is INSYS invested in medical marijuana as well, you know do they have a stake in that? And I don’t know, I didn’t dig deep enough to look into that, but yeah I think it’s, it’s always good to follow the money and look into potential motivations.

AM: That being said I also don’t want to confuse potential listeners of the podcast and have them in their minds throw this whole thing out because of sketchy funding.

JM: It is important to note that then the main findings of this study still hold true, which is that the combination of cannabis and oxycodone produces pain relief and in fact produces similar amounts of pain relief if you use less oxycodone. To reiterate, they found that the combination of cannabis and either low or high dose oxycodone was sufficient to produce pain relief even when cannabis on its own was not. Furthermore, the combination of cannabis and low dose oxycodone produced similar amounts of pain relief to the combination of cannabis and high dose oxycodone, or oxycodone on its own—high dose oxycodone that is.

BZ: But I think that is a pretty good segue into the question of like what aspects of pain are being tested here.

AM: And this is something that I also really wanted to talk about is chronic pain versus acute pain. This cold pressor test only measures one type of pain, but I know that opioids are commonly—I mean they even say in the paper that it’s commonly prescribed for chronic pain. If people can reduce the amount of opioids they need for chronic pain by using marijuana, then over the long-term use that they’ll have to use this opioid for, because of their chronic pain, it really would be beneficial in the long run and I think that’s an important part of this paper.

BZ: I think it’s important to note that the cold pressor test is actually an accepted model with predictive capability for chronic pain, even though it is actually a test of acute pain. There have been other studies which I haven’t read and I haven’t analyzed, but if we are going to trust the authors on this and you know the this field of research, they have chosen the cold pressor test as a test that seems to have predictive power over whether these results will extend to chronic pain.

JM: The key sort of next steps for testing this in chronic pain would be to actually do a clinical study and recruit chronic pain patients. You would have to probably limit it to a particular kind of chronic pain because there are many different chronic pain conditions which may respond differently. But for a preliminary clinical study, you would want to recruit chronic pain patients, but have different groups of patients either take oxycodone alone or cannabis alone or oxycodone with cannabis, and report various metrics throughout their daily lives of the of chronic pain. And that would be a step towards seeing if this really does alleviate chronic pain as compared to opioids alone.

BZ: So I think it would be useful to have a follow-up study to this that is like maybe a one-off on a single day people with chronic pain, and then like also you’d have to have a large sample size for that. But also it would be cool to have a longitudinal study following groups of patients who are taking different drug combinations over a certain amount of time.

JM: If this is longitudinal I mean you could also if they’re—if they’re so concerned about like abuse liability, you could also have more realistic metrics of that.

JM: Like you were saying Jamie, I think it’d be really interesting if you wanted to do abuse liability, give people like a big bag of medical marijuana and then prescribe a regimen for one group and don’t prescribe a regimen for another and see who sticks to it and how much they use. Overall, this is a really promising direction in research which is looking at the combined effect of cannabis and opioids to reduce the amount of opioids needed for pain management. It has really promising results showing that you can potentially have lower doses of opioids be effective when combined with cannabis. Personally, what I think this paper set out to do that it failed to do was measure abuse liability, because it ended up only comparing the abuse conditions of oxycodone to a placebo of oxycodone, and not really showing any of compelling results that cannabis had an effect on that. And same for cannabis: it seemed to compare cannabis to a placebo cannabis without showing any compelling interaction with the oxycodone. I think the authors somewhat misrepresent an interaction between the drugs in increasing an abuse liability. I think that that is an area that deserves definitely more research.

AM: I really enjoyed reading this paper. It really taught me, first of all a lot of new definitions which is always good. And I feel like this has a really good potential for chronic pain and people who need to be on opioids for long-term, and reducing that overall necessity of a higher dose of long-term opioids. Did we discuss the long-term effects of opioids and why we want to lower the effective dose?

BZ: Not really no.

AM: So the long-term effects of opioids are not great. You can have a dependence on opioids over time. Your body adjusts to having them in your system and you can need a higher and higher dose as time goes on, as your body is getting used to having that in your system. A bad side effect of that is that if you ever need to not be on that opioid anymore, your body is going to go through withdrawals because it has been used to that chemical being in its system and using it as part of your biological process. So when you remove it, now your body is going “what what happened,” and you start having withdrawal effects. And this can be mild withdrawal effects, but it can also be very severe withdrawal effects.

JM: Interestingly and kind of ironically long-term use of opioids can actually cause paradoxical pain. Which just means, if you use opioids for a long time it will actually cause you to have pain that you otherwise would not have had.

AM: So in other words if we can create a system where people need to use less opioids over that long term of chronic pain, it will be more beneficial to the person who is having to take that opioid. I really enjoyed reading this paper. I learned a lot a lot of new definitions and I think that this has a very good potential for chronic pain in the future and for reducing long-term opioid use, which has been shown to have negative effects on people. And to be able to lower that effective dose and still have the same analgesic effects, I think this is going to go in a really positive direction for a lot of people who have a lot of pain, and I really hope they continue to do this research continue to do more controlled studies. I really hope that this continues on and can be used in a really positive way to help people with pain. Thank you Jamie so much. I had a wonderful time being on this podcast and I can’t wait for the next one.

JM: Thanks hope to have you back.

BZ: I also wanted to say thank you Jamie for inviting me to be on this podcast and including me in this conversation. I learned a lot from being part of it, both from you and from Alex, and it’s always a good exercise to read these papers and really try to understand something outside your own wheelhouse. So I appreciated that opportunity and I hope to I get to do it again.

JM: I hope so too. Yeah thank you both for being on this first episode of In Plain English. I’m glad you both enjoyed the experience. I enjoyed presenting this paper. And once again we’ve been reading the paper called “Impact of Co-Administration of Oxycodone and Smoked Cannabis on Analgesia and Abuse Liability.” The paper’s authors are Ziva Cooper, Gillinder Betty, Divya Ramesh, Rebecca Balter, Sandra D. Comer and Margaret Haney. Thank you for joining us for the first ever episode of In Plain English. My name is Jamie Moffa and my guests for today have been Brendan Ziebarth and Alex Marshall. Remember to visit us on our website at inplainenglishpod.org, and give us a follow on Facebook and Twitter @PlainEnglishSci that’s P-L-A-I-N-E-N-G-L-I-S-H-S-C-I. We’ll see you next time for another paper presented in Plain English.

[Outro music]