cannabis not help psych symptoms

A recent, extensive systematic review and meta-analysis evaluated published, unpublished, and ongoing studies from 1980-2018 through several registries to assess the data on the effect of different cannabinoids on several different mental health conditions, finding scarce evidence of benefit (see marijuana not help mental disorders lancetpsych2019 in dropbox, or doi.org/10.1016/S2215-0366(19)30375-X ).

Details:
-- a detailed search found 83 eligible studies (40 RCTs, n= 3067) for the following conditions:
    -- 42 studies for depression (23 RCTs, n= 2551)
    -- 31 for anxiety (17 RCTs, n= 605)
    -- 8 for Tourette syndrome (2 RCTs, n= 36)
    -- 3 for ADHD (1 RCT, n= 30)
    -- 12 for PTSD (1 RCT, n= 10)
    -- 11 for psychosis (6 RCTs, n= 281)
-- they assessed the following cannabinoids: medicinal cannabis (any part of the cannabis plant and plant material), and pharmaceutical cannabinoids (tetrahydrocannabinol, THC; and cannabidiol, CBD) which refers to pharmaceutical grade medicinal extracts with defined and standardized THC with or without CBD content

Results:
-- pharmaceutical THC (with or without CBD) was associated with:
    -- improved anxiety symptoms among individuals with other medical conditions (primarily chronic noncancer pain and multiple sclerosis): standardized mean difference (SMD) -0.25 (-0.49 to -0.01), 7 studies (n=252), quality of evidence was very low
        -- 2 studies examined the effect of CBD on people with social anxiety and did not find any significant improvement.
        --No RCTs examine the impact of medicinal cannabis on anxiety outcomes
    -- worsened negative symptoms of psychosis: SMD 0.36 (0.10-0.62), one study (n= 24). There was no significant change in positive symptoms
    -- no significant effect on any of the other primary outcomes
    -- but adverse events, as compared to placebo, were twice as high, OR 1.99 (1.20-3.29), 10 studies (n=1495); about 3 times the withdrawal rates due to adverse events, OR 2.78 (1.59-4.86), 11 studies (n=1621)
    -- the calculated number-needed-to-harm was that one additional participant would experience an adverse event for every 7 treated with pharmaceutical THC-CBT; one additional participant would withdraw because of an adverse event for every 14 treated
-- few trials assessed medicinal cannabis, most examined pharmaceutical THC (most commonly nabiximols and nabilone), except for those that evaluated psychosis which primarily examined pharmaceutical CBD

Commentary:
-- mental health, in one small study, was the second most common reason for prescribing medicinal cannabinoids at 15%, after chronic noncancer pain in 53% (see Lucas P. Int J Drug Policy 2017; 42: 30); this small study also found that 32% of patients were able to substitute cannabis for opioids, 16% for benzos, and 31% for antidepressants [impressive changes, though this was not an RCT and subjective; though i suspect that cannabinoids are safer than opioids or benzos, even if this is a placebo effect....]
-- medicinal cannabinoids have been reported to improve all of the above mental health conditions, often through surveys, with some studies, as in last study above
    -- this publication tries to sort through the literature and provide a more evidence-based review
    --and, it is important to note that marijuana does have significant adverse psych effects, including psychosis: see http://gmodestmedblogs.blogspot.com/2014/03/marijuana-psych-effects.html
-- in terms of the specific RCTs, most were quite small, with a median sample sizes of 10 to 39 participants, and with short follow-up periods, median trial length 4 to 5 weeks
    -- and, in terms of these really small studies:
        --there was large heterogeneity in terms of methods and outcomes
        --meta-analyses are more likely to be fraught, since they are combining lots of very small apples with perhaps even smaller oranges
        --they note that “small study sizes are of particular concern as effects have been identified to be larger in small studies of medicinal cannabinoids for chronic non-cancer pain”
    -- and, in terms of short-term outcomes:
        --there might be very different risks and benefits if looking at mental health outcomes over a longer period of time (these mental health issues typically require very long-term therapy)
        --for example, it is clear that young adults (who are more likely to have some of these mental health issues) who are daily cannabis users are at risk for developing dependence
-- depression had the most studies, the quality of evidence was considered very low common in part since none of these studies included participants with a primary diagnosis of depression and most included people with other conditions such as chronic pain and multiple sclerosis

-- limitations of this analysis and with the studies available:
    -- in general, the individual studies were very small and with short-term interventions, and few studies had the mental health diagnosis as a primary endpoint (in fact participants in most of the studies did not even have an identified mental health disorder). And none of the RCTs had depression or anxiety as a primary outcome
    -- all of the studies had either low or very low quality of evidence, the only 2 exceptions were in psychosis studies, which reached moderate quality evidence
    -- it is hard to have a truly randomized controlled trial with a placebo, given the significant overall effects of cannabinoids (I suspect most participants were able to figure out whether they were on the active ingredient or not). This might be even more true for those who had used cannabinoids recreationally and were randomized to cannabinoids over placebo, who might have noted the euphoric effects more readily and also have less likelihood of adverse effects
    --there might be significant confounding in the studies on anxiety or depression, since this was not a primary entry criterion (mostly was non-cancer pain or MS), or primary outcome: eg, did the anxiety get a little better in the above analysis because the pain decreased in the group on cannabinoids??
    -- and, in the current climate of increasing availability of medicinal cannabinoids, it is unlikely that drug companies for example would fund more detailed or expansive studies; though there are 16 studies in the pipeline specifically for CBD, which had few studies done (and may be looking for a lucrative niche in the industry), including 7 for psychosis, the others in conditions where CBD has not been tested
--that all being said, this was a very impressive and thorough evaluation of published and unpublished studies (eg, this one had 6 RCTs for psychosis, vs 2 in a previous review)

So, the bottom line:
--cannabinoids are being increasingly used and available for medicinal purposes
--mental health indications seem to be among the most common reasons for prescribing them
--there really are pretty terrible studies done on the effect of cannabinoids on mental health problems, as per the above, and they do show essentially no benefit (the one exception is for anxiety, but there was a small clinical benefit, and the studies were not done on anxious people who were randomized to cannabinoids; they were studies of chronic pain control with anxiety as a side issue)

--so, as clinicians, it seems to be quite a stretch to approve or recommend cannabinoids for mental health reasons, with really no documented benefit and pretty clear risks… And, if someone is on cannabinoids (medically-justified or not), they should be monitored carefully for adverse effects

geoff​

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@uphams.org

  

to get access to all of the blogs:

1. go to http://gmodestmedblogs.blogspot.com/ to see them in reverse chronological order

2. click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

3. or you can just click on the magnifying glass on top right, then  type in a name in the search box and get all the blogs with that name in them

please feel free to circulate this to others. also, if you send me their emails, i can add them to the list