Are Americans Toking Themselves Sick?

Medscape

a day ago

This transcript has been edited for clarity.
Welcome to Impact Factor , your weekly dose of commentary on a new medical study. I'm Dr F. Perry Wilson from the Yale School of Medicine.
Whenever you see a headline that says something is 'on the rise,' there are basically two possibilities. One, the thing is actually on the rise — like pickleball.
The other possibility is that we're capturing the information differently nowadays. Either we are getting better at measuring something that has always been common (autism spectrum disorder may fall in this category, at least for part of the observed increase) or we are measuring something worse with more false positives driving up the observed rate, like UFO sightings. Ubiquitous phone cameras mean every little bit of swamp gas from a weather balloon trapped in a thermal pocket and reflecting the light from Venus gets recorded.
It's the job of epidemiologists to figure out what is driving observed changes in our health, and this week I want to share some data that show that the rate of a specific cannabis-associated problem is not just on the rise; it is absolutely skyrocketing. Let's see if we can figure out why.
I'm talking about cannabis hyperemesis syndrome (CHS), a once-rare condition characterized by cyclic episodes of what can be intractable vomiting. The exact physiology of CHS is still being worked out, but it probably has something to do with chronic exposure downregulating cannabinoid receptors in the gut, leading to decreased gut motility.
Regardless of how exactly this works, the numbers suggest it is becoming dramatically more common — at least among adolescents, as highlighted in this research letter appearing in JAMA Network Open.
Researchers interrogated the Pediatric Health Information System database which collates data from 52 free-standing children's hospitals around the United States. They were looking for adolescents admitted to the emergency department (ED) with diagnostic codes consistent with CHS, typically a chief complaint of nausea and vomiting, and at least a second diagnostic code indicating a 'cannabis-related concern.' These could be things like cannabis abuse or dependence. The time frame of interest was 2016 to 2023.
Let's just take a look at the raw numbers. To give you some perspective, in 2006, there were 6.8 ED visits for CHS per 1 million population. In 2016, the start of this study, that number had jumped to 160.4, a more than 23-fold increase over a decade. These rates increased around 40% per year through 2023. The most recent data showed around 2000 visits for CHS per million population; that's a 30,000% increase in less than 20 years. Crazy numbers.
But… are they real?
Remember, when we see the rate of anything going up, we need to make sure we're not measuring it differently. And I certainly have some concern here that we are potentially unmasking a problem that has been there all along but not well documented until recently.
Adolescents show up in the ED all the time for nausea and vomiting. The AHRQ publishes some of these data. For example, in 2018 there were nearly 800,000 visits for nausea and vomiting among those under age 18 — about 2.5% of all ED visits.
The question is, how many of those had CHS? The study we're discussing this week would call it CHS if, in addition to the nausea and vomiting, they had a secondary code for a cannabis-related issue. I have no doubt that we use those codes more frequently now. For one thing, the broad-scale decriminalization and legalization of marijuana has removed much of the stigma that existed in the early 2000s. Doctors might be less worried about 'outing' these kids nowadays. Another possibility is that knowledge about CHS is spreading; with more marijuana use, doctors are getting more sensitive to the diagnosis. So perhaps some of those things we used to call a 'stomach bug' now get appropriately diagnosed as CHS.
The converse could also be true. Since we're more comfortable adding cannabis-associated diagnosis codes now than we used to be in the past, we may currently be misclassifying people with nausea from another cause as CHS. Just because someone has cannabis use disorder or cannabis dependence doesn't mean they are not allowed to get norovirus.
Of course, the other possibility is that there is a lot more marijuana use going on and with that, more CHS. The authors hypothesized that, if that were the case, we might see a more dramatic rise in states that have legalized recreational marijuana use.
This is where the data get a bit weird. Because, yes, states with recreational marijuana legalization had more ED visits for CHS — overall, about 1900 cases per million vs 800 cases per million. But the rate of growth of those visits was more dramatic in states without recreational marijuana legalization.
The authors don't opine on what would cause this pattern of observations. That's probably wise since the data are relatively limited. But that has never stopped me before, so here is what I think is going on.
I think a lot of this is real. I suspect there is a true increase in the amount of CHS that is commensurate with the broader availability of marijuana in the marketplace. This leads to higher rates in states that have legalized recreational marijuana use.
But a significant proportion of the observed increase is due to uncovering CHS in people who would not have received the diagnosis in the past and misclassification of non-CHS syndromes in the setting of people who happen to use marijuana.
This occurs in states without legalized marijuana because, let's face it: Even in those states, the stigma about marijuana use is nowhere near what it was 10 or 20 years ago. The times they are a-changin', as one former pothead noted.
Is this all a problem? CHS can be bad. In fact, just under half of the kids in this study required a hospital admission; a bit under 1% required the intensive care unit. That said, we should still contextualize these ED visits in the context of other risks adolescents face. I pulled a bit of data comparing ED visits for CHS to those for depression, suicide, gun violence, and drug overdoses.
You can see here that CHS rates aren't as high as, say, ED visits for suicidal behavior, but these are all in the same ballpark. Of course, that's assuming the numbers in this study are not inflated. Still, it seems like we can add CHS to the list of real risks adolescents face in the United States today. And, if growth rates of the syndrome continue as they did in this study, rising by 50% per year, well, we can assume that every American adolescent will be in the ER for CHS by 2040.
That would be… unlikely. But I suppose we can keep a bloodshot eye on it.