Yale Psychiatry Grand Rounds: December 17, 2021

December 17, 2021

Information

"The Epidemiology of Cannabis Use/Disorder in the United States"

Deborah Hasin, PhD, Professor, Columbia University

ID7301

To CiteDCA Citation Guide

00:00To introduce Doctor Deborah Hasan.
00:02Dr Hasan is professor of epidemiology
00:05at Columbia University in the
00:08Department of Psychiatry and in the
00:10Mailman School of Public Health,
00:12the Department of Epidemiology.
00:14Doctor Haasis research has covered
00:17epidemiological and clinical studies of
00:19substance use and substance use disorders,
00:21particularly alcohol and cannabis and on the
00:25comorbidity associated with these conditions.
00:27Her studies have had a
00:29substantial impact on the DSM.
00:31Five definitions of substance use disorders,
00:34and she was recently text editor
00:36for the revised version of DSM 5.
00:39She's also leveraged epidemiological
00:41data to answer important questions
00:43for clinical research and practice.
00:46In this regard,
00:47I've really had the pleasure to to work
00:49with her on her research validating
00:51non abstinent alcohol reduction as an
00:54outcome for use in treatment studies
00:56and her work has been pivotal in
00:58helping move this agenda forward.
01:01By any metric, Dr.
01:03Hasan has been extremely productive.
01:05She's been a principal investigator
01:07of naida and anti AAA grants since
01:091990 and published over 500 papers.
01:1324 of her first author papers have
01:15been cited over 100 times and two
01:17of them over 1000 times showing
01:20considerable impact on the field.
01:23She's also committed to training
01:24and mentorship.
01:25She directs the night up funded pre
01:27and post doctoral training program and
01:29substance use epidemiology at Columbia
01:32and many of her mentors have gone on
01:33to successful careers of their own.
01:35Receiving NIH and other funding is
01:38faculty members at numerous universities.
01:41Now, with respect to today's presentation,
01:43her work and the work of her trainees has
01:46contributed to a better understanding
01:48of cannabis use and the effects of
01:51state cannabis laws across the lifespan.
01:53This issue, as you know,
01:54is very timely for Connecticut,
01:55where possession of 1.5 ounces
01:57of cannabis is now legal.
01:59As of July 1.
02:00Today we will learn more about changing
02:03trends in cannabis use nationally,
02:05and what the impact of medical
02:08and recreational cannabis laws
02:09are on rates of cannabis use.
02:11Cannabis use disorders and the use
02:13of highly potent cannabis products.
02:16So with this brief introduction,
02:17let me welcome Dr.
02:19Hasan as today's grand round speaker
02:21and turn the floor over to her.
02:23Thank you.
02:26Well, thank you Stephanie for for
02:28that kind introduction and thank
02:30you all very much for inviting
02:32me to give this grand rounds.
02:34So I guess the next challenge
02:36is correctly sharing the screen.
02:45Do people see what looks like
02:47the first slide of a grand rounds
02:51of a PowerPoint? Put it in
02:53slideshow mode for yourself right there.
02:56OK great excellent.
02:57So so I'm going to talk about several
03:01different things related to cannabis
03:03use and cannabis use disorder,
03:06and I'll talk you through some of
03:09the thinking that I do about this.
03:12Just focused on cannabis and also
03:14in relation to other substances like
03:16for example alcohol and opioids that
03:19have had such huge ups and downs.
03:21Over time I'm going to talk a little
03:23bit about cannabis use disorder.
03:24I'm going to talk about national
03:26time trends and.
03:28Teens and adolescents and trends in
03:30adults 'cause these are important to
03:33know in and of themselves and also to
03:36to understand the effects or lack of
03:40effects of of the changing cannabis laws.
03:43So I will get started with that.
03:45Let's see if I can do this.
03:46Yes, OK,
03:47so I'd like to start out by acknowledging.
03:51My grant support that
03:53contributed to this talk,
03:55and the fact that I have
03:56research funding for an unrelated
03:58project from Syneos health.
04:02OK, so cannabis is I think most people
04:04that are on on you know listening
04:07today would know that cannabis is one
04:09of the most widely used psychoactive
04:11substances in the US and worldwide.
04:14Of course many people can use
04:16cannabis without any particular harm.
04:21Oops. But use is associated with the
04:25risk for numerous health conditions,
04:27including acute, cognitive,
04:29and motor impairments,
04:31withdrawal, withdrawal syndrome,
04:33respiratory problems,
04:35vehicle crashes in people
04:37that start using early,
04:39you know, early in adolescence,
04:41there have been a lot of long term
04:43prospective studies that have suggested
04:45that early use does create long term
04:48problems and functioning and achievement.
04:51It's a cannabis use is a necessary
04:54condition for cannabis use.
04:55Disorder and cannabis use disorder
04:57is also associated with substantial
04:59comorbidity and impairment.
05:00As we'll see in a few minutes,
05:02so due to all these risks,
05:03understanding the changing
05:05epidemiology of cannabis is important.
05:09So shifts in US attitudes and rates of
05:12use of different substances are not new.
05:16A Yale historian David Musto
05:18was writing about this in the
05:21early 1990s as shifting attitudes
05:24pertained to alcohol and to opioids.
05:28And specifically to cannabis,
05:29there have been ongoing debates for a
05:32long time about the extent to which
05:34cannabis use and cannabis policies
05:36have harmful effects on users tended
05:39to polarize into people who minimize
05:41potential harmful effects of cannabis,
05:44and argue strongly for more
05:46permissive or liberalized laws,
05:48and those who continue to view
05:50cannabis as a harmful drug,
05:51for which continued prohibition
05:52is the correct approach,
05:54I would say at this point continued
05:56prohibition is not strongly
05:58endorsed by most people.
06:00But that does represent the sort of.
06:03Alternative camp to people that
06:05think that the law should just
06:07be changed to be liberalized.
06:11And I'm going to bring up an example
06:12from the past that I'm going to come
06:14back to at the end of this talks.
06:16So of course everyone knows
06:18about alcohol and prohibition.
06:20You know, alcohol wasn't accepted substance,
06:23then it wasn't.
06:24Then it was prohibited across the whole US,
06:27and then it was brought back and we
06:29are living through the tag end of a
06:32process where opioids prescription
06:33opioids were seen as positive,
06:35used incredibly widely,
06:37and then we've seen the terrible
06:39consequences of that through the open.
06:42The opioid overdose epidemic.
06:45Well,
06:45a little bit more recent example
06:48that I think it's useful in thinking
06:51about the cannabis laws is a
06:53debate and policy shift around a
06:55minimum legal drinking age.
06:57So in the late 60s in the early 70s,
07:00this was a state experiment, if you will,
07:03in states lowering their ages,
07:04most of them to around age 18
07:06on the grounds of fairness,
07:08the thinking was well,
07:10if people can vote at age 18,
07:12it's only fair to let them
07:14purchase alcohol legally.
07:15Also,
07:16but then what happened is many studies
07:19were conducted that showed that these
07:21lower minimum legal drinking ages
07:24predicted many adverse consequences.
07:26So in 1984,
07:27by the time a lot of this evidence
07:30had accumulated,
07:31a federal act was passed that required
07:34a uniform age across all states,
07:36which was age 21,
07:38and that was put into place by 1986.
07:40So I'll come back to this one
07:42at the end of the talk,
07:44when we look at what's going on with
07:47cannabis state cannabis laws at this point.
07:49So, uh, specifically with respect to
07:52cannabis, I think most people know
07:54there's been there have been huge
07:56changes in how people viewed it.
07:57So in the 1800s marijuana was actually
08:00used for many medical purposes,
08:02but it's used for those reasons
08:04became less important with the
08:06development of synthetic painkillers.
08:08There were there,
08:10the public view was actually
08:11transformed in the 1930s after Mexican
08:14immigrants introduced recreational
08:16marijuana and anti marijuana and anti
08:19immigrant views sometimes got blended.
08:21There was a 1936 article in the
08:25American Journal of Nursing.
08:27Voice to very extreme view
08:30of marijuana users.
08:31Probably many people have seen on YouTube
08:35the the the old movie clip video,
08:39Reefer madness and public opinion
08:42has really fluctuated markedly
08:44overtime with respect to cannabis.
08:46I think it's clear that we're going
08:48through an era where the attitudes
08:50are changing to the positive.
08:51If you look across the beginning
08:53and end of this graph,
08:55you'll see that Americans have increasingly
08:59supported legalization of cannabis.
09:01And this is only to 2018,
09:03I think if we did,
09:05if if the study were redone now,
09:07it would probably be an even higher
09:09percent and I'll come back to this this
09:12map a little bit later in in the talk.
09:16But as we see at this point.
09:20Medical cannabis is legal in 37 states,
09:22so states are in the minority at
09:24this point that haven't legalized
09:26any form of cannabis use,
09:27and recreational cannabis is legal for
09:30adults in 18 States and Washington DC.
09:34OK,
09:34so I'm going to talk just a bit
09:36about cannabis use disorder criteria
09:38because they come into play and
09:40thinking about some of what we
09:42know about cannabis use disorder.
09:44So here you see the DSM 4 criteria.
09:48Those were changed in 2013,
09:50which is actually getting to be awhile ago.
09:52It's eight years ago at this point
09:54and here are the criteria for DSM 5
09:58and you can see that specifically with
10:01respect to cannabis, there were two.
10:04New criteria that were added withdrawal.
10:07There is a specific cannabis withdrawal
10:10syndrome and craving for cannabis.
10:12Craving was added to all the substance
10:15use disorders including for cannabis.
10:18So cannabis use disorder in
10:20cannabis users is a necessary
10:23but not sufficient cause of CUD.
10:25There is a common assumption or
10:27at this point you might even say a
10:29myth that very few users of cannabis
10:32develop cannabis use disorder,
10:34and that's based on a paper from
10:36the early 1990s that suggested that
10:38cannabis use disorder was rare.
10:41But think back,
10:42this is during a time when the
10:44potency of cannabis that was used
10:45was very mild and the diagnostic
10:47criteria were quite different.
10:49At that point also,
10:50so with changing norms and more
10:53potent forms of cannabis,
10:54this this concept is outdated,
10:57so there was a meta analysis that
11:00was published pretty recently that
11:03showed across different studies that
11:06the estimates of rates of cannabis
11:09use disorder amongst users is in
11:12the range of 2020 or so percent.
11:15These numbers get higher if you
11:17look among more frequent users.
11:20Then I checked them out in some more recent
11:23data than what we're available to the.
11:25The people that conducted this meta
11:27analysis and these numbers hold up,
11:29so it's not necessarily the case that
11:32it's fine for everybody to use cannabis,
11:34because nobody is going to
11:36get cannabis use disorder.
11:38So just to look at some descriptive
11:41epidemiology of the prevalence
11:42of cannabis use disorder,
11:44this is these these.
11:46This information is from the
11:47National Epidemiologic Survey on
11:49alcohol and related conditions.
11:51This was a national survey that
11:54was conducted in 2012 and 2013.
11:57It was a probability sample
11:59with over 36,000 Americans.
12:02The measures in this study were
12:06tested extensively and it's
12:07still a good source of data.
12:09On a lot of conditions that aren't
12:11available in other surveys,
12:13So what you see here is at
12:15that time in adults,
12:17the overall prevalence of cannabis
12:19use disorder was about 2.5.
12:22And disaggregated by mild,
12:25moderate and severe levels
12:27according to the DSM,
12:29five definitions about half
12:30the cases were at the mild
12:32level and a quarter each at the
12:35moderate and the severe level.
12:39This shows the risk for cannabis use
12:42disorder by demographic characteristics.
12:45We see here that there was
12:47a greater risk in males.
12:49There's a greater risk in African
12:52Americans and Native Americans,
12:54especially at the severe end,
12:57and if you look at the bottom of this table,
12:59you see that younger people were
13:01more likely to have cannabis
13:04use disorder than older people.
13:07The study also shows very clearly
13:10that cannabis use disorder is
13:12highly and highly associated with
13:14other substance use disorders
13:16and with psychiatric disorders.
13:21And this shows that impairment and
13:23functioning which is at least half a
13:26standard deviation between the below
13:28the population norm on a standardized
13:31scale was common among people that had
13:34cannabis use disorder and increased as
13:37we went across the the levels from mild,
13:40moderate to severe.
13:41So cannabis use disorder is a real condition.
13:45It does really exist,
13:46it uses the same criteria as
13:48other substance use disorders.
13:50And it's associated with comorbidity
13:52and impairment and then in a paper that
13:55came out just recently done by one of
13:57the fellows in my training program.
13:59We looked at the relationship
14:01between having cannabis use disorder
14:03and not having any substance use
14:05disorder where the whole set of other
14:07problems that people could have.
14:09And this shows the relationship.
14:11So across a whole raft are of
14:13different types of problems,
14:15interpersonal problems,
14:16financial and legal problems,
14:18and health related conditions.
14:20Those with cannabis use disorder
14:22were significantly more likely and
14:25substantially more likely in many cases.
14:27To have these problems compared to people
14:30without a substance use disorder at all.
14:33So cannabis use disorder is a
14:35condition to take seriously.
14:37So I'm going to shift now to
14:39looking at some trends,
14:40overtime in cannabis, and I'm going to.
14:42I will talk first about adolescents
14:44because adolescents and adults
14:46do actually tend to go their own
14:48way in terms of substance use,
14:50epidemiology.
14:51So I'll start out by showing you
14:54some information from monitoring
14:55the future monitoring.
14:57The future is a yearly survey
14:59that's been conducted every year
15:02and 12th graders and since 1991
15:04and 8th and 10th graders and the
15:07amazing thing that's very huge.
15:10Advantage of this study is that they've
15:12kept their questions constant overtime.
15:13The basic questions on perceived
15:15risk and use of the substances,
15:18so it's possible to get very long term
15:20perspectives from monitoring the future data.
15:22On what's going on with trends
15:25in in students in 8th, 10th,
15:29and 12th grade.
15:30So what we see here is there's been
15:32some fluctuation.
15:33Overtime in perception of great
15:37risk associated with using cannabis,
15:40but it's got it's pretty much gone
15:43down so consistent with what we'll
15:45discuss a little bit later in adults.
15:48Also,
15:48the perception that it's harmful to
15:51use this substance has declined.
15:52Usually that's associated with
15:55increased use of a substance,
15:57although it hasn't been so closely
16:00entwined to that in teens in recent years.
16:03So here's some information on
16:06long term time trends in any use
16:09of cannabis over the past year,
16:12and as you can see,
16:13in 12th graders it's been pretty
16:15flat for the last several years.
16:17There's been some increase in the last
16:20couple of years for Ace and 10th graders.
16:24But they are not as marked as they
16:26as increases that we're seeing
16:28in in the beginning of the 1990s.
16:33And here we see
16:33daily use and we see pretty similar patterns.
16:37Not too much evidence of increases in
16:40the 12th graders, but some increases
16:43in the 8th and the 10th graders.
16:46And some worrisome trends in a
16:49paper that was published last year.
16:51Frequent use among lifetime users,
16:54so this show this figure shows over the
16:59years male female differences in monitoring
17:03the future participants in frequent use.
17:07You can see that it's
17:09gone up in recent years.
17:12And this is in white participants
17:14and in black participants who
17:16actually have gone up more sharply
17:19and ended up at a somewhat higher
17:21place than the white participants.
17:23So we don't see huge change overall
17:26in in any use of cannabis weed,
17:28but we do see some trends in heavier
17:31use of cannabis amongst adolescents.
17:34Well, what about adults?
17:37So taking a look at two surveys that Nice
17:42arc survey that was like a predecessor
17:45of the Nice arc three that I showed you
17:48earlier that was conducted in 2001,
17:502002 and about 10 years later,
17:53the Nice arc three was conducted and
17:55pretty much the same questions and sample
17:57design were utilized in both surveys,
18:00enabling people to look across
18:01the the surveys to see about
18:04increases in rates overtime.
18:06So this is past year cannabis use.
18:09In adults at least 1818 years and older,
18:12and you can see that the
18:14rates more than doubled.
18:16And this is cannabis use disorder
18:18where we also see an increase in
18:21rates nearly doubling across.
18:23You know, across the two surveys.
18:26And,
18:26as shown in the paper,
18:28if you're interested in taking a look,
18:30any use in cannabis use disorder
18:33did increase significantly within
18:34major demographic subgroups,
18:36including men and women within age groups,
18:39income and race ethnicity groups.
18:44But there were there were data that came out
18:47of a different series of national surveys.
18:49The National Survey on Drug Use and Health.
18:52I'm sorry about the typo
18:53in the in the title there.
18:55So Wilson Compton,
18:56who is now Deputy director of Naida,
18:59published a paper about five years ago
19:01showing a lot of changes overtime in
19:05cannabis use attitudes towards cannabis
19:07and then cannabis use disorders.
19:10So the thing that was kind of
19:12striking about this paper.
19:13Was that although a lot of other
19:15things were changing in the news?
19:17Do data.
19:18These are nationally representative
19:20surveys that are conducted each year.
19:22Is that the rates of cannabis use
19:25disorder didn't change overtime.
19:27They were about 1.5%.
19:29And that was true across the years
19:32of this survey and an additional
19:34analysis of these data were conducted
19:37and published that showed pretty
19:39much the same thing,
19:40so it wasn't so clear why.
19:43Denise Arc was showing increases,
19:46but the news do was showing that
19:48although use was increasing and
19:50frequent use was increasing in adults,
19:53the rates of cannabis use disorder
19:54were staying the same.
19:56So this was an epidemiologic
19:58debate for a while.
20:00And then there was a new paper
20:02that came out where the team pulled
20:04in a couple of additional years
20:05of data, and although the DSM five
20:08criteria weren't utilized in those years,
20:12in the Miss do surveys,
20:13they created a proxy of the the DSM 5
20:18diagnosis and that when they use that
20:21one they did show increases overtime
20:23and the increases were largely in
20:25the mild category and they considered
20:27the DSM five to be a more sensitive.
20:30Indicator, but for this paper a
20:32caveat was that DSM 5 craving and
20:35withdrawal were missing in the news.
20:37Do so it was still unclear how the results
20:39would have looked if these were included.
20:42Well, very recently than is do
20:45has published 2020 data where
20:47they did shift to DSM 5 criteria.
20:51So I pulled these numbers together
20:53for this presentation.
20:54So we see putting 2019 and 2020 data
20:58together just first looking at cannabis.
21:01Use any cannabis use in the
21:03past year across the age groups.
21:06We don't see any any evidence
21:08of a huge shift there.
21:09It doesn't look like the the
21:11prevalence of use went up.
21:13In the young people or the mid
21:16you know the the young adults or
21:18or the adults aged 26 and older.
21:21However when we look at the
21:24shift from DSM 4 to DSM five,
21:27we do see quite a difference.
21:29We don't see that much of a
21:31difference in the 12 to 17 year olds,
21:33but in the young adults we see
21:35quite a large difference from DSM.
21:37Four to DSM five there were a lot of things.
21:40Of course that happened in
21:422020 including the.
21:43The pandemic and a lot of other things.
21:45But if we just think about the shift
21:48from a set of criteria that was used
21:50for a long time and then is due to the
21:52new criteria that were expected from
21:54the beginning to be more sensitive,
21:56we see a big increase that looks
21:58like it could be artifactual in
22:00the young adults and also in the
22:03in the adults age 26 and older.
22:05So as an epidemiologist this just means
22:08that it's hard to use new data over
22:11long periods of time to look at trends.
22:13And if we want to have a deeper understanding
22:16of what trends are going on nationally,
22:19we need to look for other sources of data.
22:21And that's important not only to understand
22:23what's going on in this country with
22:25cannabis use and cannabis use disorders,
22:27but also to understanding
22:29the effects of the laws.
22:31So one other source of data
22:33includes cannabis use,
22:36disorder diagnosis and
22:38inpatient hospitalizations,
22:40and we can see that in this
22:42paper between 2002 and 2000.
22:4511 There was an increase of
22:47about over doubling of the rates.
22:49Now,
22:50that doesn't mean that the people were
22:52hospitalized for cannabis use disorder,
22:54but they.
22:55These were one of the diagnosis
22:56that were made.
22:57If they were in in inpatient hospitalization.
23:02And in a paper that updated the
23:05findings to somewhat more recent years,
23:07these this paper looked over a
23:10longer span of time and showed
23:12quite a substantial increase using
23:15pretty much the
23:16same variable. Now the authors
23:18of this paper said that they
23:20didn't shift to looking at ICD
23:2210CM that occurred in 2015,
23:262016 because that presented too
23:29many methodological challenges.
23:30But I will talk about those in some data.
23:33That we have to address that
23:34in a in a minute or two.
23:38One of another fellow from my training
23:41program, a postdoctoral fellow,
23:42looked at the prevalence of past non
23:45cannabis use in pregnant and non
23:48pregnant women of reproductive age.
23:50And the overall finding was that
23:52the rates of cannabis use were
23:54lower in the pregnant women,
23:56but the rates have changed over the period
23:58of time addressed with Miss do data,
24:01we're pretty much the same with
24:04increases in pregnant women as
24:06well as non pregnant women.
24:09And this is another take on a similar issue.
24:13This paper just came out very recently.
24:16We weren't involved in this paper,
24:17but you can see that in in women
24:21people that were hospitalized
24:23for pregnancy or pre pregnancy,
24:27the rates of cannabis use disorder went up,
24:30including importantly cannabis use
24:32disorder being the only substance
24:34use disorder diagnosis which is
24:36indicated here by the little.
24:39Gold colored circles.
24:44So this was a study that was
24:47published actually actually at
24:48this point a number of years ago,
24:50and this study actually was something
24:52that led us to getting a grant from
24:54Niger that I'll be talking about in a few
24:57minutes to look more closely at cannabis
24:59use disorders across time in the VA.
25:02But this paper, which went from 2002 to 2009,
25:06showed an increase of 59% across
25:09this time in cannabis use diagnosis.
25:13And working with the VA data across
25:16a longer period of time with
25:19the grant funding that we got,
25:21we looked at this including the
25:25transition period from ICD 9 to ICD 10.
25:28So the first several years in this show,
25:32the changes overtime in ICD 9 CM and
25:35what we see here is a clear increase
25:38overtime in all the groups rates were
25:42highest in the youngest veteran patients.
25:45This was the point in time when the
25:48criteria were changed in the VA
25:50and we see that there's a decrease
25:53right after that change.
25:54That was,
25:55do we think to some administrative
25:58issues in recording old diagnosis
26:01versus new diagnosis,
26:03but we see that once that period
26:05was passed that we once again see
26:07a pretty steady increase overtime
26:09in the prevalence of cannabis use
26:12disorders across the five and a
26:14half million or so.
26:16Patients that were treated in
26:18the VA each year.
26:20This is how the similar findings looked when
26:22we broke them apart by males and females.
26:24This is the period with ICD 9 CM.
26:27This is the transition point and this
26:30is the post transition point findings
26:33where you see the rates were going up
26:36pretty steadily in males and females.
26:39And this is what we get what we
26:41got when we looked at this by race.
26:43Ethnic groups once again with
26:46the two periods.
26:48This paper is,
26:49we've got a paper on this that
26:52we're just about ready to submit.
26:54That reports on these findings,
26:57so it seems pretty clear looking across
27:00survey data and across information from.
27:06Large electronic medical record data
27:08that cannabis use disorder has been
27:10increasing in the United States.
27:12Use has increased in cannabis use.
27:14Disorder has increased as well,
27:16so then it makes some sense to think
27:19about what are potential factors
27:21that are going on in in that.
27:24So
27:25I'm going to talk about
27:27increases in in cannabis potency.
27:29I'm going to talk about increases in
27:31pain and then I'm going to spend a fair
27:34amount of time talking about increasingly
27:37permissive state cannabis laws.
27:39So this this slide shows the
27:42changing potency of illicit cannabis
27:45samples that were seized by the Drug
27:49Enforcement Administration across time.
27:51And as we can see,
27:53there's been a quite a substantial increase
27:56in the potency of the seized samples
27:58going up to two 2014 in this figure.
28:05And then continuing the findings further
28:08into a few additional years up to 2019.
28:12So there's quite an increase in the
28:15potency of the product that's being
28:17used and potency is thought to be
28:21related to addiction liability of
28:23different substances that are used.
28:26But now, in addition,
28:27we not only have the illicit
28:30cannabis products that people have
28:33been able to get for a long time,
28:34but we have legal cannabis markets and
28:37a lot more products that people can use.
28:41So we have plant or flower products,
28:45and at this point the average potency
28:48in those products is from 15 to 20%.
28:50That's including legal ones as well
28:53as illegal ones. We also have.
28:56Changing products in terms of concentrates
28:59concentrates that can be used.
29:02It can be vaped or or dabed,
29:05and those are really quite
29:08concentrated and potent.
29:09And.
29:11And are increasingly preferred
29:14by cannabis users overtime and.
29:18It it appears from some different
29:21surveys that people are seeking out the
29:23more concentrated forms because they
29:25they like various aspects of the effects.
29:28So I'm going to let that sit
29:30for a minute or two.
29:31I'll come back to potent
29:32forms of cannabis later.
29:36What about pain?
29:38So pain could be considered an emerging
29:41risk factor for cannabis use disorder.
29:44There's somewhat mixed evidence on the
29:46efficacy of cannabis to treat chronic pain,
29:49as shown in two large reviews.
29:53And it would be better to have
29:55more rigorous studies of this,
29:56of course, but Americans anyway
29:58are increasingly seeing cannabis as
30:01an effective treatment for pain.
30:03A colleague of mine at the
30:05VA in San Francisco,
30:07Salma Kahani has a paper showing this and
30:10with changing US attitudes towards cannabis,
30:13is pain and emerging risk factor for
30:15frequent use and cannabis use disorder.
30:18Well, we know that the prevalence of
30:22pain is increasing in US adults from a
30:25paper that was published awhile ago.
30:28So we undertook to look at the
30:32relationship of pain to frequent use of
30:35cannabis and cannabis use disorders.
30:37In these two surveys,
30:39the Nice arc that was conducted
30:41at the beginning of the Odds 2001,
30:442002 and the Nice arc three that
30:47was conducted about 10 years later.
30:49So here the green lines represent
30:52the participants who did have pain.
30:55Is measured with a widely used pain
30:58question and the blue ones are the
31:02participants that didn't have pain.
31:04So if we look at the earlier one first,
31:07we see that there's not much difference
31:10in the prevalence of frequent non
31:12medical cannabis use in the Nice
31:14arc by whether the participants had
31:16pain or not and if we look in the
31:18nice arc three we see that there
31:20is a difference and that difference
31:23was actually significant.
31:24And when we looked at whether the overall
31:27results differed between the two surveys,
31:29we found that they did.
31:32So that's frequent use that's
31:34daily or near daily use.
31:37And then we did the same thing
31:39for cannabis use disorder.
31:40And here we saw that there was a
31:43significant difference between those with
31:44pain and without pain in the earlier study.
31:49There was a greater difference
31:53in the more recent survey.
31:56And once again, we found that there
31:58was a significant difference between
32:00the surveys with stronger relationships
32:02shown in in in the more recent study,
32:06which I think a lot more research
32:08needs to be done on this.
32:10But it does suggest that having pain
32:12is a risk factor for non medical
32:15cannabis use and cannabis use disorder.
32:19OK, so now I'm going to shift to looking
32:22at various aspects of changing state
32:24cannabis laws and how they may have
32:27affected cannabis use cannabis use disorder,
32:29and maybe some other consequences also.
32:33So we've addressed this in the
32:36grants that we've had, which we had
32:39one to study this in adolescence,
32:42and we have now the one to
32:43study it in adults in the VA.
32:45We've used a multi level framework,
32:47so we look at the national picture.
32:50Then we look at state level factors,
32:52and then we're also looking at
32:54individual level factors as well.
32:56So the national picture hasn't
32:58really changed in terms of legality
33:00for a really long time since 1970,
33:03and marijuana is still right up there with
33:06heroin as being a schedule one substance.
33:09Which may not make sense to people,
33:10but that's the way it's been for a long time,
33:13and changing these things is in at the
33:16federal level is really a huge challenge.
33:19But we do know that other national
33:21aspects of cannabis use from the
33:23slides that I've heard you show
33:25increasing rates across time of use.
33:30But one factor that has also been changing
33:33that can be addressed are state level laws,
33:35and that's what I'll be talking about more.
33:38So first we can think about how could more
33:42relaxed cannabis laws affect cannabis use?
33:46The medical laws there's there's two
33:48factors that are commonly assumed in
33:50the theoretical literature to increase
33:52use of a psychoactive substance.
33:55One is the desirability of the substance,
33:58and one is its availability.
34:00So medical cannabis laws could
34:03increase desirability by signaling
34:05that cannabis use is safe.
34:07And also could increase availability
34:10by dispensaries or home cultivation.
34:13Recreational cannabis laws could also
34:16convey the idea that cannabis use is safe.
34:19It's also no longer illegal,
34:21so if there's some people that are
34:22refraining from using cannabis because
34:24of the fact that it's been illegal,
34:26if the state law changes they they
34:28may change their mind about that.
34:30Certainly commercialization and
34:32advertising would be permitted if the
34:35state law that was passed allowed that.
34:37There's there's some variation
34:39in what could be allowed,
34:40but generally they do permit advertising
34:42and all of these things could reduce
34:45disapproval and perceived harmfulness,
34:47and obviously increased availability would
34:50be achieved via retail outlets outlets.
34:54The fact that no medical authorization is
34:57needed and people can do home cultivation.
35:00So I'd like you to look at this map.
35:03Once again,
35:05this shows how widespread the
35:07the laws have become.
35:10Cannabis was the very California was the
35:13very first state to legalize medical
35:15use back in 1996 and the West Coast
35:20states all followed pretty quickly.
35:23The first recreational cannabis
35:26law was passed in 2012.
35:29Those two states,
35:30the two states that passed
35:32were Colorado and Washington.
35:34There are no states that have leaped
35:37right to recreational cannabis
35:39use without having a period with
35:42medical cannabis laws in place.
35:46So I'm going to talk just a little
35:48bit about some complications to
35:50studying state medical cannabis laws
35:53and recreational cannabis laws,
35:55so of course they're not randomly assigned.
35:58There are people that talk about studies
36:01of these state laws as natural experiments,
36:05but they're really not because
36:07they're not randomly assigned,
36:08so that has to get taken into account,
36:10and also states changed their
36:12laws in different years,
36:13so it wasn't like one year we had
36:15states with no laws in the next year
36:16we had states with different laws.
36:18So the rates have been changing
36:20overtime states have been changing
36:22their laws in different years overtime,
36:25and it's also the case that when states
36:28pass the medical or recreational laws,
36:30they may already have higher rates
36:32of cannabis use due to other factors
36:35like socio demographic characteristics
36:36of their population or attitudes
36:39and their population.
36:40So we need study designs that
36:42will address these issues.
36:43If we can do any better than just
36:47looking at association of the.
36:49State law with the risk of use which
36:52associations are not necessarily don't
36:55necessarily tell us anything about causation.
36:58So a common analytic strategy that's
37:00been used in most of the studies
37:02I'm going to show you is it's called
37:04difference in difference tests,
37:05and So what these tests look at is
37:07does the average change in states
37:09before and after that change in
37:11a law differ from the average
37:14contemporaneous change in states
37:15that don't change their laws?
37:18And so we can see based on the fact
37:22that use has been changing over time.
37:23In adults it if we look at states
37:26that don't change their laws and
37:27consider them a can a control group,
37:30you would see a trend overtime.
37:31But there's no law that changed in those
37:34states. So it's just a solid line.
37:37And on top we have the states that
37:39did change their laws overtime.
37:41So if the laws had no effect,
37:44what you would see is a parallel trend
37:47from the left part to the right part.
37:49As shown by the extension in the dashed line.
37:53But if there is an effect of the laws,
37:55you would see that by having a break
37:57in that trend on average with with
38:00a greater change overtime in the
38:03states that changed their laws.
38:05So adolescents are a particular concern
38:08because we know that cannabis use
38:11may permanently impair neurocognitive
38:14functioning.
38:15Questions about that are being addressed
38:17at this point in the ABCD study,
38:20but.
38:22With Hnyda funded several studies
38:23a few years back to specifically
38:26investigate the relationship of changing
38:28cannabis laws to adolescent use.
38:31So we published an early paper Melanie
38:34Wallace, a statistician in our group,
38:35and she took the lead on this paper where
38:39we looked at the relationship of cannabis,
38:42the differences in prevalence of youth,
38:44use of cannabis in states with.
38:47Medical cannabis laws and states
38:50without them and here are the rates
38:52in the non non medical cannabis law
38:55states across several years of data.
38:58And here are the rates in the
38:59states that did have them so you
39:01can see that on average the rates
39:03are higher in the states that did
39:05have these laws and the differences
39:07are all statistically significant.
39:09But as Melanie wrote in the paper,
39:11the cause was unclear because association
39:14doesn't necessarily show causality.
39:18And then we were funded by NAIDA
39:20to use monitoring the future data
39:22to examine this using difference
39:23in difference approaches.
39:24So we started here by just looking at
39:28the simple associations once again.
39:30We see here that this is the
39:33difference between the ever States
39:34and the NEVER States and we see
39:37that the rates are lower in the net
39:39than ever states significantly.
39:40So in most cases.
39:45But when we did, the difference
39:47in difference look across time.
39:48We didn't show that rates were higher after
39:52passage of the medical cannabis laws.
39:54So you can see here that there's really
39:57no evidence of increases in any of these
40:01groups combined or separated by age.
40:03In fact, there was a decrease in the
40:05eighth graders that we didn't expect,
40:07but we interrogated that result
40:09quite a bit statistically,
40:11and we couldn't make it go away.
40:12So it seemed to be pretty robust.
40:15And this is this is the result
40:17that we got in this very large data
40:21set from monitoring the future.
40:23So we did this study and then many
40:26other people did studies of the same
40:29issue using monitoring the future data.
40:32And other datasets too.
40:33As you can see here,
40:35than is due and and others as well.
40:37So errands are vet who was in
40:40our group at the time,
40:41did a meta analysis of these
40:44different studies and showed
40:46aggregating across all the results.
40:48There was really no post passage
40:50increase in cannabis use in teens.
40:52Instead,
40:53in studies of them that used appropriate
40:56difference in difference tests.
40:58And a paper was recently published
41:00that extended this by a couple of more
41:03years and also looked at whether the
41:05states permitted dispensaries or not,
41:08and when and whether that made a
41:10difference and what we see here.
41:12These odds reissues are all
41:13very close to one,
41:14and none of them are greater than
41:17one that indicates once again that
41:19the laws didn't were not associated
41:21or not predictive of increases in
41:24marijuana use across the youth.
41:26In this study.
41:28So in summary,
41:30it doesn't look like the laws have
41:32encouraged greater use of cannabis
41:34in in people in the adolescent ages.
41:38But what about in adults?
41:39So let's take a look at what
41:42that picture looks like.
41:43So this is a study cross sectional
41:48association study that we did when we
41:50first started examining the the picture,
41:52and we see that in non MCL States
41:55and MCL states the rates were higher
41:58in the MCL States and they were
42:01significantly so controlling for a
42:03whole bunch of different factors.
42:06There was a study,
42:07an early study that was done by
42:09these authors using Miss Do data
42:11that I think is worth pointing out.
42:14And this is use of cannabis across time,
42:17so the the vertical bar in the
42:20middle is the date when the.
42:23Law was passed in that state and to
42:26the left is looking before passage and
42:29to the right is looking after passage.
42:31So there were no significant increases
42:34in these states on average in adult
42:37use before passage of the laws.
42:40But there was an increase after
42:42passage of the law and that that
42:45increase was maintained and this is
42:47in use past month cannabis use and
42:50then the authors also looked at.
42:53Posed the same question for DSM
42:55four cannabis use disorder and they
42:58did this for using the same year.
43:00The law was passed and then using one in
43:03two year lagged analysis on the grounds
43:05that it could take a year or two for
43:08people to develop cannabis use disorder.
43:10If if they've started using it and they
43:13found that using these lagged analysis
43:16that medical cannabis effects medical
43:19cannabis law effects were also found
43:21in adults for cannabis use. Disorder.
43:25So working with Silvia Martins in
43:29our group at Columbia, we looked at.
43:33We used this do data,
43:36updating it by a couple of years
43:38and also looking across age groups
43:41because of the age groups,
43:43the differences that were becoming
43:44pretty well known by that point.
43:46So what we see here is that looking
43:49at the prevalence and non medical law
43:52States and medical law states that
43:55there are no significant differences.
43:58By by the status of the law in
44:01the youth or in the young adults,
44:04but with the difference in difference tests,
44:06we see that there were differences
44:08in the adults across across the
44:11adult ages 26 and older,
44:13and breaking it down by specific age groups.
44:16We see that this finding holds across all
44:19these different age groups in adults.
44:23So this was using this new data and
44:25that went back as far as 2004 to 2013.
44:27One of the things I wanted to do was to
44:30be able to look across a greater span
44:33of time because I was interested in
44:35seeing what would happen if we looked.
44:38If we had survey data that could be
44:41used that that existed before any of
44:43these medical laws had been passed
44:45and see what that would look like.
44:48So the nice arc and the nice arc three
44:50I've already talked to you about.
44:51There was an earlier study about 10
44:54years earlier that used the same substance,
44:56use and substance use disorder
44:58measures in the same sample design.
45:00Pretty similar sample design,
45:02so the end lays the National longitudinal
45:05alcohol and associated disorders.
45:07I don't even remember the acronym anywhere.
45:10Study was conducted in 1991, nineteen 92,
45:13and that was a point when there
45:15were no medical cannabis laws.
45:17As I mentioned,
45:19cannabis California passed their
45:20first law in 1996 and then in the Nice
45:23ARC at the time that that was done,
45:26there were several states that had
45:28passed their laws and in the Nice arc
45:30three there were several more states.
45:33So what we did was we looked in the
45:35earlier period and then we looked
45:37in the later period and in this and
45:40a few additional figures I'm going
45:42to show you the red lines are always
45:44the states that didn't change their
45:46laws or pass a medical marijuana
45:48or medical cannabis law during the
45:51period under consideration.
45:53And California we pulled out because
45:55it always behaved differently from
45:57the other states,
45:58so we can see thinking back to
46:00this earlier period that may seem
46:02kind of foreign now because,
46:04but it was a period when cannabis
46:06use was going down.
46:07Actually not going up,
46:09we saw parallel decreases in California
46:13and the states that didn't change their laws,
46:16but in the five states that did change
46:19their laws between 1991 ninety two to 2001,
46:222002, there was a significant significant.
46:25Increase in use and we saw the same
46:28thing for cannabis use disorder.
46:31In the more recent period,
46:33there were more things going on,
46:34so this is a little bit
46:36more complicated slide.
46:37But once again, the red bar here shows
46:40you the states that didn't change
46:43their laws during this time period.
46:45The gold bar shows the states
46:48that were late passing states.
46:50Nine states passed medical cannabis laws
46:52between the Nice arc and the Nice arc.
46:55Three this is where they started out.
46:57This is where they ended up and you
46:59can see that the rate of increase
47:01if you will was faster in the
47:03states that did pass those laws.
47:05This is what happened with California.
47:08And this is what happened with Colorado,
47:10which ended up at the top of the pack.
47:12Actually,
47:13even though it started out
47:15pretty close to the.
47:16The other states at the beginning
47:19of the time period.
47:20And this is what we got when we
47:22look at cannabis use disorder.
47:23The number of participants with
47:25cannabis use disorder was lower,
47:27so it was harder to achieve
47:29statistically significant results.
47:30But we see something pretty similar
47:32in terms of the patterns for
47:35never changing their laws during
47:37the period to the late states.
47:40The changes for California.
47:42And then changes for Colorado.
47:44So this suggested that something
47:47was going on.
47:48Putting this together with the news
47:51do findings and suggested that the
47:53medical cannabis laws were affecting
47:55the risk for cannabis use and cannabis
47:57use disorders in adults 18 and older.
48:02So I'm going to turn now to recreational
48:05cannabis use and the laws permitting it,
48:08so the benefits of these laws are, you know,
48:11there are people that really genuinely
48:13believe that they're important benefits
48:15to be gained and and these these are.
48:17You can't really question these too much.
48:19OK, to eliminate unfair racially
48:21based arrests, that's for sure.
48:23And also to generate jobs,
48:25business income tax revenues and
48:27also to give voters what they want
48:30when they're passed by the ballot.
48:32Expectations amongst people that have
48:34been looking at this area for a long time
48:37is that it will expand the pool of users.
48:39These laws there where there's
48:43commercialization permitted that there
48:44will be people in there that will be
48:47very interested in expanding the pool
48:48of users and the laws if they permit
48:52it will increase commercialization's
48:54more availability,
48:55more advertising,
48:57potentially affecting desirability and
49:00possible kind of sneaky efforts to attract.
49:03Underage users and this could
49:06potentially lead to a number of
49:09adverse outcomes shown here.
49:13So studies on the effects of recreational
49:16cannabis laws are really just getting going.
49:19This is a paper that we did with magnet
49:22Serta who now is at NYU and we looked at.
49:25We use difference in difference
49:27tests to look at the effects of
49:30recreational cannabis laws across
49:32the ages and what you see here.
49:34The odds ratios,
49:36the adjusted odds ratios that are in bold,
49:39or the ones that are significant.
49:40There's one there for.
49:42Cannabis use disorder in
49:44the youngest age group,
49:45but when we did sensitivity analysis
49:48it they suggested that this was
49:51due to unmeasured confounding,
49:53so we took more seriously.
49:55The findings in the 26 plus age group which
49:58were robust to this type of confounding,
50:02and you see that for any cannabis use
50:05frequent use and cannabis use disorder.
50:07The risk was increased after passage
50:10of the recreational cannabis.
50:12Laws that were in existence at the
50:15years encompassed by the survey data.
50:20OK, so some of the last data that I'm
50:23going to show you here pertains to
50:26MCLR CL in the use of potent products.
50:30So if we wait for a national survey data
50:33to give us the information for this,
50:35we may have to wait really a long time.
50:38And sometimes there's emerging important
50:40issues that just need more rapid data to be
50:44collected that can answer the questions.
50:46So sometimes collection of that data
50:49is actually pretty straightforward.
50:50Although there can be caveats raged
50:53raised about generalizability and one such
50:56method is online social media survey data.
51:00So on on in work for a project that I'm
51:04doing with Alan Budney who's at Dartmouth
51:09we're doing work to try to develop a
51:12more accurate measure of cannabis exposure,
51:15cannabis use and as part of that
51:17we're analyzing a number of different
51:19datasets as well as creating a measure
51:22that we're going to be testing out.
51:24So one of the datasets that we've
51:26analyzed is from a social media
51:28survey that was conducted in 2017.
51:31And this was reaching out to
51:34people via Facebook advertising's.
51:35It's a very inexpensive way to get
51:38a very large sample pretty quickly,
51:40and so this survey included over
51:434000 adults in 50 states in DC,
51:46and we use the data from this
51:48study to look at the relationship
51:50of recreational cannabis laws,
51:52medical cannabis laws,
51:53two states without these laws in 2017
51:56to the use of highly potent cannabis
51:58products with concentrates that I showed you.
52:01Earlier, so this is association not
52:04difference in difference on causality.
52:07But there isn't so much information
52:09about this topic,
52:10so we wanted to add to it.
52:12So, descriptively, this is what we got.
52:14We looked across the know
52:16cannabis law states,
52:17MCL States and RCL States and you
52:20see a pretty regular increase in the
52:23the proportion of participants that
52:25use the potent cannabis products.
52:28And when we subjected this to analysis,
52:30controlling for a lot of individual
52:32level and state level variables,
52:35we found that recreational laws compared
52:37to states without such laws was in
52:40significant significantly increase the
52:42risk of using the concentrates and
52:45when we did a three level analysis we
52:48saw that both recreational and medical
52:51cannabis laws increase the risk and
52:54that the the resident difference between
52:56the recreational and medical cannabis law.
52:58Effects where working with data now
53:01to look at this in a survey that
53:04was that was just conducted in 2021,
53:06but I don't have the data yet in
53:09a form that can be presented.
53:12So to summarize, teen Cannabis users siudi.
53:16We haven't seen any evidence of
53:18substantial national increases
53:19in the last several years,
53:21despite the fact that perceived
53:23risk has decreased in young people.
53:26And there's been this proliferation
53:28of states that have legalized use for
53:31medical and recreational purposes.
53:33We see that use is higher in
53:35states that ever pass these laws,
53:37and the lack of pre post differences in
53:39when the you know in passage of the laws.
53:42Suggest that cannabis legalization
53:45is not encouraging adolescent use,
53:47but of course for a topic like this.
53:49Continued research and
53:51monitoring is very important.
53:54What about adults? Well,
53:55I think we've seen that cannabis use is
53:59associated with considerable problems,
54:01impairment and comorbidity.
54:02I'd like to emphasize again that the risk
54:05of cannabis use disorder amongst cannabis
54:08users is greater than commonly assumed,
54:10and there's very abundant evidence
54:12that the prevalence of cannabis
54:14use and cannabis use disorder have
54:16increased in 2000 since 2000.
54:17In EU. S. Adults.
54:20In terms of the cannabis legalization,
54:24I haven't talked about this at all in
54:26this talk because it would take too long,
54:28but there's been suggestions that
54:30legalizing cannabis might solve.
54:32The opioid crisis based on ecological levels.
54:35Studies that are that are somewhat
54:38flawed for addressing the question in
54:41individual level studies of those with pain,
54:44cannabis users are actually more
54:46likely to use opioids, not less.
54:49Non medical cannabis users.
54:51But we see from the data that are
54:53available that the state medical
54:55cannabis laws increase the risk for
54:57adult cannabis use and cannabis
54:59use disorders and the state
55:01recreational laws do as well.
55:02We don't have as much data yet,
55:04so obviously more information is
55:07needed and we are actually working
55:09on that with the VA data right now.
55:12So I'm going to end with
55:14this slide on implications.
55:16You know,
55:16I began working in this area
55:18with a public health perspective,
55:20which is to provide information about the
55:23effects of these laws and what they might.
55:26You know what they might be doing
55:28in terms of public health problems.
55:31So we have already the the status that
55:3337 states have passed the medical
55:36laws 18 have recreational laws.
55:38Many more states are considering
55:40these laws at this point,
55:42and it doesn't seem that considerations
55:44such as cannabis use disorder are.
55:46Really,
55:46a high priority in the debates
55:49about Legalization II.
55:51Don't hear them being discussed much at all.
55:53But I will say coming back to where
55:56we started that there have been other
55:58long term shifts in substance attitudes,
56:01policies and prevalence.
56:02It's part of American life.
56:04Thinking back to what I talked
56:06about at the beginning in terms of
56:08state level shifts from the best
56:10of intentions and the minimum legal
56:12drinking age.
56:13But then the role of evidence in
56:17shifting those,
56:19the changes in the drinking age
56:21back to age 21 to avoid.
56:24The problems that occurred by
56:26lowering the age,
56:27so I sort of see my function at this
56:29point with this type of research as
56:31serving a public health function
56:33to continue to monitor and provide
56:36information to policymakers to
56:37provide medical and health providers
56:39and to the public,
56:41it can ultimately inform policy
56:44and protect health.
56:46And with that I think I'm out of time
56:48and I'll stop and I thank you very
56:50much for inviting me to speak here.
56:54Thank you so much Debbie.
56:56That was terrific. There are some.