|Click to hear Stacey read this page.||
The most common concern raised in cannabis legalisation discussions is potential harm to adolescents. There are two main avenues of concern, the first being cognitive deficits (For instance lower IQ or poor educational outcome) and psychosis (eg schizophrenia).
Studies of these factors are widely conflicting. Research is impeded by the difficulties involved in studying a population with widely varying health, social, economic, and other drug use for the effects of an uncertain dose of a widely varying complex herb, with hundreds of biologically active chemicals which can all vary in their concentration. Potential adulterants and contaminants in street cannabis don't help clear the air much either.
Major confounding factors that could be involved are concurrent tobacco/alcohol/drug use, genetic factors, socio-economic status, history of abuse, and several genetic factors.
Some difficulty arises because no certain and known pathway
of schizophrenia for instance has been discovered. Several environmental
and genetic scenarios have been proposed. Different studies have
shown completely conflicting reports. For instance one study
will show conclusively that COMT gene is to blame, and no other:
then another will prove just as indisputably that COMT is not
involved at all and AKT1 is the culprit.
Risk factors for schizophrenia or psychosis symptoms previously identified among scientific literature include:
Mother's health, nutrition and stress during pregnancy
Most of these factors are directly related to or at least more common in those of low socio economic status, which in itself tends to increase the incidence of early cannabis use.
|Links between ethnic identification, cannabis use and dependence, and life outcomes in a New Zealand birth cohort||
The educational outcome has also been compared to ethnic and socio economic status in a New Zealand study which shows links between ethnic identification, cannabis use and dependence, and life outcomes in a New Zealand birth cohort. The results show that although initially it looked like Maori youth were at higher risk for both cannabis use and poor educational outcomes, when the results were controlled for socio-economic status and family function, the incidence of Maori cannabis use was no longer significant.
When cannabis use alone was controlled for, the Maori ratio of educational outcome was also not significantly changed. Socio economic and family/home environment were not controlled for, and probably accounted for the majority of the difference to judge from the other studies in similar matters overseas. Other substance use, including tobacco and alcohol were not investigated or controlled for either, and are usually higher in cannabis-using teenagers.
|Can Poverty Drive You Mad? 'Schizophrenia', Socio-Economic Status and the Case for Primary Prevention||
The potential for socio economic standing to contribute to schizophrenia, consistent with the discrepancy and conflicts between studies purporting to prove genetic risk factors for schizophrenia and their association with environmental factors like cannabis has been addressed in an excellent paper by John Read of the Auckland university, prepared for the Hunter address Poverty and Schizophrenia 2011, where he explains in detail the evidence linking low socio economic status to a approximately fourfold increase in schizophrenia cases.
I highly recommend reading the paper, which is linked to at the left. The relationship between 'schizophrenia' and poverty was described as 'one of the most consistent findings in the field of psychiatric epidemiology'. Socio economic status was related to more variables (such as other environmental factors, or symptoms) than any other variable involved.
In addition the poor were more likely to be hospitalised, were kept in hospital longer and received fewer treatments at discharge. They tended to receive physical or pharmaceutical treatments, where the higher socio economic classes tended to receive referrals to counselling services. Low SES patients also had a higher rate of re-hospitilisation.
So socio economic standing is one reason why cannabis use could be more prevalent in populations who are also more prone to psychosis. However this doesn't confirm whether cannabis could be contributing to the cause of these disorders. However checking the facts on schizophrenia in NZ and worldwide tends to indicate that there aren't many, if any, new diagnosed cases of schizophrenia that genuinely would not have occurred if the patient had not taken cannabis.
The prevalence of cannabis use in teenagers in NZ has increased fairly steadily since its introduction by american marines from vietnam in the 60's. It is now up to at least 70% of today's teenagers overall, with more use by young men than women. Similar rates have been recorded right across the globe.
|Assessing the impact of cannabis use on trends in diagnosed schizophrenia in the United Kingdom from 1996 to 2005||
If this caused any new cases
of schizophrenia, (or for that
matter poor educational outcome such as Leaving school at or
prior to 18 without a passing grade), that would not have occurred
otherwise, there would be a proportional
increase in both new
diagnoses and prevalence overall.
The same scenario is true for poor educational outcome (such as Leaving school at or prior to 18 without a passing grade). Educational achievement figures have certainly not worsened either, (although this is harder to compare, as great efforts are made by our educational institutions to ensure better and better educations are achieved), and the average national IQ has not decreased. (It would have been expected to have decreased by about 2 points nationwide by now if the figures shown in the Dunedin study were new deficits caused by cannabis use).
|Testing hypotheses about the relationship between cannabis use and psychosis||
In analysing these prevalences, the Australian study at the left is a great read on the subject, analysing four plausible association scenarios between cannabis and schizophrenia, and how the actual live population data fitted those scenarios.
The first proposal is Does cannabis cause Schizophrenia? The expected result of this scenario is that the incidence and prevalence of schizophrenia in the Australian populations studied would have increased proportional to the increased use of cannabis. As we have explained above, the population data failed to support this hypothesis, and in fact contradicted it.
The second proposal is Does cannabis cause an earlier onset of schizophrenia that would already have occurred but later? The expected result of this scenario is that the lifetime incidence level would stay the same, and the age of onset would be lower by about 1 year in cannabis users. The population data matched in that there was no increased incidence of schizophrenia, and was inconclusive about the age at onset as one study showed the age of onset had decreased slightly, and two showed it had not. An increased incidence of new psychoses being professionally diagnosed as Drug-Induced possibly bears out the scenario, but the author mentioned this could also be caused by doctors assuming cannabis was involved causally because they had been told it could be. (In fact this could also have been other drugs involved in the diagnosis, such as amphetamines and synthetic cannabis)
The third proposal is Does cannabis use worsen the prognosis of Schizophrenia by making relapse more likely? The expected result of this scenario is that although the incidence and age at onset would remain the same (the results of which we have already addressed in the above two hypotheses), the overall Prevalence of schizophrenia in the population as a whole would increase to involve those who would have perhaps recovered otherwise in the cannabis-using population.
However, when the actual figures that would be expected were calculated, because of the existing high rate of relapse in the non-cannabis using population, the total increase in prevalence would be seen as only 1% of the total prevalence of schizophrenia. Since schizophrenia is already only 1-2% of the population, this put the expected figure on paper at 1 in 10,000 people, which was too small a number to confirm in this study population.
This is a very likely scenario however, and should not come as a surprise to anyone. Certainly an excessive dose of a very stimulating Sativa type of cannabis will cause a drastic worsening of certain psychotic symptoms such as paranoia and delusion, and this episode can lead to complete relapse in sensitive individuals. Alcohol abuse increases the likelihood of relapse too. In any case, this being later life use of cannabis causing recurrence, it is more appropriately further addressed in the Cannabis Effects for Adults page than here.
The fourth proposal is Are Schizophrenic people simply more likely to use Cannabis? The expected result of this scenario is that the incidence, prevalence and age at onset would remain the same, the ratio of cannabis users in the Schizophrenic population would be about twice that of Non-schizophrenics, and would have increased steadily at this level in proportion to the increase of use in the rest of the population. The population data matched this scenario in every way, with the exception of the uncertainty over the age at onset dynamic, which was inconclusive.
This paper more or less confirms what we already knew or could logically suppose: namely that
This last is the area where the biggest questions lie, and this is where more research would be of great value. This is a very plausible scenario, as most schizophrenics are not effected and diagnosed overnight: they struggle for years with ever-worsening symptoms before they finally are diagnosed and treated. Use of a high-anxiety Sativa type cannabis (which is the most common variety adolescents get into) would certainly make them feel a lot worse and could instigate earlier diagnosis.
The current research indicates possible earlier onset in cannabis using patients as compared to non-using, but this does not prove the cannabis caused the earlier onset. If looked at from the 4th scenario point of view, that schizophrenics were more likely to use cannabis, then this could be seen as a consistent Dose-Related response, ie the patients with the worst degree of the disorder who are diagnosed earlier being even more likely to use cannabis than those with the mildest symptoms. The only way to prove this yea or nay is to compare the average age at onset over all NZ from prior to the 1960s to today and see if it has changed to a consistent degree as cannabis use has increased.
|Evidence of a Cohort Effect for Age at Onset of Schizophrenia||
One study done in France in 2001 showed results that are not what anybody expected and are frankly concerning but don't appear to be linked to cannabis. The study was done to compare the age at onset and age at first hospitalisation between 3 cohorts of subjects: those born prior to 1944, those born 1945-64, and those born 1965-84. It was not intended to compare any drug use to the age at onset, however this factor was compared as a possible confounder. They showed that the average age at onset of schizophrenia had decreased significantly and consistently over this time, from 25 years to 20 years. However when Substance abuse history (yes or no, binary answers) was compared there was no significant difference in age at onset between those who had a history of any Substance Abuse to those that did not.
What this means is not clear and more research is needed to confirm these figures. If the increase in cannabis (and other drug) use since 1944 was to blame for the lower age at onset one would have expected this study to show the age at onset different between drug users and non-users, however this was not the case, only some 2 months of age difference were seen which is not statistically significant. Whether other factors (such as nutritional, cultural or chemical exposure) factors are involved and if so what those factors could be is not possible to suggest from this result.
Another bone of contention in the (cannabis use during adolescency vs mental disorders) debate is the potential association with child abuse. A great proportion of victims of child abuse use cannabis (and other drugs, including tobacco and alcohol), and this tends to muddy the waters around the potential causes of certain disorders.
One side of this fence says that the effected individuals' disorders were most directly associated with the history of abuse, and the cannabis is only involved as a self-medication or coping mechanism.
|Plenty of studies show links between abuse (in childhood or early adolescency) and Schizophrenia or Psychotic symptoms. Most of these studies show a good Dose-response relation, for instance the Netherlands Mental Health Survey and Incidence Study (NEMESIS)!!!, which showed a 3 to 14 times increase in the likelihood of developing psychotic symptoms in those who experienced any traumatic event such as abuse or neglect before 16. With increasing number and frequency of abuse events, that went up to 48 times the normal risk.|
One study, done in britain (actually by some of the same scientists who did the Dunedin and the Avon studies which we will address further down the page), was based on a population of same-sex twins. They showed consistently that childhood maltreatment by adults or bullying by peers was significantly linked to psychotic symptoms at age 12. The critical factor was that of Intent to Harm; an accidental trauma did not have this effect.
Lastly we have a study that shows that cannabis is not the cause in these instances and child abuse is, by comparing to each other.
A middle view sitting on top of that fence is that the two are coincidental, and both could be contributing.
|Cannabis Use and Hypomania in Young People: A Prospective Analysis||
The other extreme says that the subjects acquired their disorder entirely because they used cannabis, and that there is no direct link between child abuse and any mental disorder.
There is really only one study that seriously tried to push this last argument however. The study methods were poor and the verdict was biased.
Firstly I have to take issue with the method of data collection on child abuse. They record child abuse on the basis of the child's mother reporting it in a questionnaire she received in the mail. Now I'm aware that the majority of child abuse, particularly of the sexual variety, does not come from the child's mother. But not all. Emotional and physical abuse is common from mothers too, and is she really going to fill out a form she got in the mail to say Oh Yes I abuse my baby. Is she even going to fill it out to say My partner abuses my baby? Please. Mothers whose partners abuse their children often don't know, or they live in a state of Denial or Self-deception. Checking that Yes box would have to be the hardest thing you could ask those women to do.
The study made no attempt to verify the accuracy of this questionnaire data with the child's own report later in life. They could have asked the 18 year olds when they tested for bipolar symptoms, to confirm the accuracy of their mothers' early reports. They did not.
Further, they made no attempt to check for any child abuse after the age of 7 years. This is a problem in most girls who get sexually abused it happens around puberty. They just missed the most likely and worst time for the issue they were supposedly testing for.
After this, their conclusion is that Cannabis is the cause of the abused children becoming Bipolar later in life. Their reason for this conclusion is essentially that when they fed the basic figures (ie, the prevalence of cannabis use in bipolar victims of child abuse) into a computer program and asked it if it was possible that cannabis could be the cause, it agreed that based on those figures it could be possible. Therefore they conclude that Cannabis is the direct cause of their disorder, and the reason so many abused children are bipolar is that (for some reason that apparently remains mysterious to the study authors) they use more cannabis in adolescency than non-abused children.
The concept that these kids could be resorting to drugs to treat serious mental and emotional distress, (otherwise known as a coping mechanism) and that that distress could be the sign of a risk for diagnosable disorder, apparently never entered into their methods.
|The role of childhood trauma in bipolar disorders||
In counter to this proposed pathway, many other studies have already shown direct links between abuse and bipolar disorder. However even better than this, another review investigates the biochemical changes found in victims of child abuse, for instance an increase in brain levels of pro-inflammatory cytokines such as tumour necrosis factors. These are the chemicals responsible for destruction of damaged cells, a normal part of the immune system.
However if the levels are elevated beyond an appropriate level healthy cells can be destroyed in large numbers. These findings strongly indicate a direct correlation between child abuse and damaging biochemical changes in the brain, which are linked to Bipolar and other disorders.
Interestingly, in research into auto-immune disorders, both THC and CBD have been shown to reduce the production and action of these exact same cytokines.
One possibility for a coincidental high usage of cannabis in schizophrenic patients is in the self-medication hypothesis. To verify whether this is valid most studies have investigated symptoms beginning before cannabis use, such as psychosis symptoms at age 11. Across several studies, these adjustments reduce the potential causative effect of cannabis use, suggesting self-medication is likely to be involved to some extent, but it is still elevated.
However these studies check the easy and obvious symptoms of psychosis, and potentially miss the point: it is not only the positive psychotic symptoms (eg delusions, paranoia) that affected individuals attempt to treat with cannabis. It is the Negative symptoms, such as apathy, social withdrawal, lack of interests and feelings etc. THC-containing cannabis may relieve these symptoms. These symptoms are known to occur in schizophrenic individuals long before the positive symptoms and diagnosis, and to increase the likelihood of conversion to psychosis later in life, making evaluation of the easily-defined positive psychotic symptoms unsuitable for diagnosing the self-medication effect for this subject.
|Psychosis reactivity to cannabis use in daily life: an experience sampling study||
A study of schizophrenia patients in regard to the self-medication dynamic showed that although the timing of cannabis use did not seem to be related to their symptoms, cannabis relieved some of the symptoms of schizophrenia, and increased the subjects positive mentality and happiness scores. This suggests a simple explanation for the question of why so many schizophrenic patients continue to take cannabis despite the potential increase in paranoia and delusion. The benefit may be seen to outweigh the detrimental effect.
This is not to say that it is an appropriate medication for the condition with widely disparate patient symptoms, the effect of the different types of cannabis on this disorder is very individual. However it does say that research into medicinal cannabis for psychosis could be very enlightening on the subject. If we were evaluating the efficacy of a standard medication for these symptoms, then the prevalence we see of continued use of the medication despite potential side effects (which all the anti-psychotic medications do have, in some cases to a debilitating degree) would indicate a good rate of customer (patient??) satisfaction.
In fact, some studies have now shown that CBD has anti-psychotic effects, reducing the positive symptoms of schizophrenia. It has the opposite effect to THC in this regard.
The presence of CBD in whole-herb cannabis preparations may explain why some studies show less incidence of psychotic side effects. This counter is present in higher quantities in some varieties than others, and may negate the acute effect of THC on anxiety. Some people have found a heavy Indica strain high in the CBD side of the CBD/THC ratio to be therapeutic for positive symptoms of schizophrenia.
|Adolescent THC Exposure Causes Enduring Prefrontal Cortical Disruption of GABAergic Inhibition and Dysregulation of Sub-Cortical Dopamine Function||
One potential course of this action by CBD is seen in animal trials. After exposing adolescent rats to THC, a research team found it had reduced their GABA (an inhibitory neurotransmitter) levels and thereby caused neurons in the animals' frontal cortex to become hyperactive in adulthood. The rats then developed schizophrenia-like symptoms such as higher levels of anxiety and lower social motivation.
They could reverse these effects in the brain and in the rats' behaviour by using drugs that activate and enhance the GABA system.
They produced these results with THC alone. Using CBD in the same scenario could be interesting. CBD influences the GABA binding site (Receptor) to have a greater affinity for GABA. This results in increased GABA action as more GABA is bound to the receptors to better effect. This would suggest that CBD could be the ideal companion to THC.
Even within cannabis there may be radically different effects depending on the proportions of THC to these other components. For instance a Sativa strain is typically very energising and stimulates creative thought: the side effect of this is in that is may provoke paranoia in vulnerable individuals. An Indica strain on the other hand is very calming and sedative. The primary difference is that Sativa is high in THC and low in CBD, an Indica has higher levels of CBD as well. Other cannabinoids and terpenes are also altered in the different varieties which accentuate the disparate effects even further.
Functional Magnetic Resonance Imaging (fMRI) have confirmed that while pure THC can increase psychotic symptoms in an acute administration period, pure CBD has the exact opposite effect. THC has very active and stimulating properties: CBD has very sedative and calming properties. CBD is essentially that antidote to the side effects of pure THC. However it doesn't stop there there are at least 60 different cannabinoids present in cannabis, plus hundreds of terpenes and similar chemicals. These all work together to accentuate some effects and counter others.
For this reason when discussing the potential psychotic impacts
of this herb, it is important to distinguish between pure THC,
synthetic cannabinoids, and cannabis herb or whole-herb extracts.
Synthetic cannabis has a reputation for being Far more potent than the natural herb. This is because it has a greater Affinity for the CB1 receptor than THC does. Basically, it is a Stronger drug than even purified THC.
As another confusing factor, Synthetic cannabinoids, these pure chemicals, are generally what is used in animal studies in laboratories, for legal reasons. Some studies do use purified THC for this purpose, but even this cannot be considered to be the same as the whole herb.
Synthetic cannabinoids, including some that were not supposed to be psychoactive, were developed for the relief of pain and nausea in cancer patients, but most of these were discontinued because of the unacceptably high incidence of psychosis-type side effects. This is an example of how single cannabinoids, particularly synthetic ones, can have more detrimental side effects than the whole herb with the natural synergy in place.
Synthetics contain no CBD (one of the antidotes to psychotic
symptoms found in cannabis) and none of the other valuable phytochemicals
Synthetic cannabis has been strongly linked to acute psychotic symptoms lasting for a week or more. Whether permanent effects may be linked to this substance has not been well researched at this point. However by its nature as a potent THC-type molecule without any CBD-type accompanying it, it has the greatest potential for harm of any of the cannabinoid drugs. Therefore it is very relevant to the subject and should have been included in the studies as a reasonably separable criteria.
|Moderation of the Effect of Adolescent-Onset Cannabis Use on Adult Psychosis by a Functional Polymorphism in the Catechol-O-Methyltransferase Gene: Longitudinal Evidence of a Gene X Environment Interaction||
One of the best studies on schizophrenia and cannabis-use in adolescency was done here, the Dunedin multidisciplinary health and development study, who published a paper on cannabis use in adolescence and the adult onset of Psychosis (a more general term for disorders like schizophrenia, the official diagnosis of which is not a very accurate variable to use in a scientific setting). In this study they showed a significant potential link between cannabis use in adolescency and psychotic symptoms at age 26.
They found a significant association with one gene in particular, namely the Catechol-O-Methyl-Transferase (COMT) gene. This gene is involved with a dopamine deactivator, and varying levels of this gene result in either high or low dopamine levels, which are linked to mental disorders. This would be consistent with other studies that have showed cannabis lowering dopamine levels in relatives of psychotic patients, where it raises the dopamine levels of others. Unfortunately many studies since this have failed to show this gene association, and no exciting breakthroughs have yet come from the genetic discovery in this paper. In fact other studies since have conflicted even with the findings of the association of psychosis with adolescent cannabis use.
One methodological issue with this study is also that they only adjusted if Adult tobacco and alcohol use affected the results. No adolescent use of tobacco was mentioned at all. Nicotine has also been shown to increase dopamine expression in schizophrenic individuals.
Many teenagers who smoke cannabis also smoke tobacco cigarettes. Even more commonly they mix tobacco with their cannabis when smoking it, making it very hard to evaluate the level of nicotine exposure.
|Does tobacco use cause psychosis? Systematic review and Meta-analysis||
A review in 2014 shows the evidence for nicotine exposure being associated with higher risk and earlier onset of schizophrenia. The association shown is one very similar to that seen in cannabis. This does not mean it is a causal relationship but it does suggest the importance of verifying the concurrent level of tobacco use in the adolescents in these studies.
This review indicates that tobacco could effect schizophrenia by increasing the sensitivity of dopamine 2 receptors. Excessive dopamine activity has been strongly linked to schizophrenia. Interestingly, one of the proposed methods by which THC is proposed to effect schizophrenia is by increasing the release of dopamine (THC may not effect or may even reduce dopamine in most individuals. However at least one study has shown it is increased in relatives of psychosis patients).
If this proves to be the case, then tobacco and cannabis taken together or at least by the same person could have a synergistic effect, where by working together they have greater effect than both individually. In this case an extra factor of risk involved with polysubstance use needs to be investigated and controlled for.
|Tobacco is even more likely to be involved with the cognitive deficit involvement, as tobacco use in adolescence has been linked to lower IQ levels. The Avon Longitudinal Study of Parents and Children shows that although initially the use of cannabis in adolescence seemed to result in lowered IQ scores, when the teenagers use of tobacco was accounted for, the association with cannabis disappeared entirely. The same teenagers tended to use tobacco, and tobacco did strongly impair mature IQ levels.|
|Persistent cannabis users show neuropsychological decline from childhood to midlife||
The Dunedin study publication on cannabis and IQ scores contradicts this result.
The adolescent cannabis users had lower IQ scores in maturity, and removing tobacco from the picture didn't help at all. Unfortunately we're not Apples for Apples to the Avon study above as they do not remove the confounding effect of tobacco in the same way. They use Persistent Tobacco Dependence as their measure, not Ever-use as the Avon study did.
This Dependence is defined as dependence at at least three evaluations. In fact only three evaluations had ever checked for tobacco dependence criteria in the dunedin study therefore we can be sure that these people rated as Persistent Tobacco Dependence were dependent on tobacco at the ages of 21, 26 and 38. Unfortunately this does not help us define if these same subjects used tobacco during adolescence. They probably did but lots of other kids (including most of the cannabis smokers) also would have and not shown on the Persistently dependent scale later. 91% of the teenagers today who use cannabis mix tobacco with their cannabis on a regular basis.
Most of these kids seldom actually smoke a Tobacco Cigarette. When asked if they smoke cigarettes, these kids' first answer is typically No, as they do not think of themselves as tobacco smokers, and they certainly do not meet Tobacco Dependence criteria. Yet they consume a large quantity of tobacco over time, mixed with their cannabis, and eventually this will out in conversation. The prevalence of tobacco consumption by the same adolescents who used cannabis regularly has not been publicly reported in the dunedin cohort as yet.
However, this study differs quite drastically from many others in that removing the subjects who rated as dependent on tobacco for so long did not reduce the IQ decline shown at all. The total lack of IQ harm from tobacco for instance is inconsistent with all other studies so far. These individuals Should have been high on the list of individuals who did use the most tobacco in adolescence too, suggesting that for some reason the dunedin cohort did not show IQ damage from tobacco where most other studies have.
Also in a reply to a challenge on Socio-Economic confounders, the author declares that low socio-economic kids were no more likely to use cannabis. This is not consistent with any other study on the subject either, which all show low socio-economic status to be strongly associated with a greater likelihood of using cannabis. It is also not consistent with what we as New Zealanders see on the streets. Verification of this data would be nice.
|Associations between adolescent cannabis use and neuropsychological decline a longitudinal co-twin control study||
More recently several of the same authors published a new work produced from another study in britain comparing cannabis-using Twins to their non-cannabis using paired Twin. In this they prove the opposite result to their previous work in dunedin, saying Family background factors explain why adolescent cannabis users perform worse on IQ and executive function tests.
Initially similar to the dunedin study the results (taken over the whole population the same as if they were singletons) were of deficits in IQ and cognition. But when cannabis users were individually compared to their own twins, these results disappeared. The authors conclude Short-term cannabis use in adolescence does not appear to cause IQ decline or impair executive functions, even when cannabis use reaches the level of dependence. Family background factors explain why adolescent cannabis users perform worse on IQ and executive function tests.
|Impact of adolescent marijuana use on intelligence: Results from two longitudinal twin studies||
Other studies have been done comparing cannabis using/non-using twins to each other too.
One was particularly comparable to the dunedin study and showed clearly that the neurocognitive decline was not caused by cannabis.
This study of twins showed exactly the same result as the dunedin cohort on initial analysis, namely that the IQ of cannabis using teens deteriorated over the period between 13 and 18, where non-cannabis using teens did not.
However, again when the individuals were compared to their non-using twins, no statistically significant differences were seen, and notably the slight differences that were seen were just as likely to be positive as negative. On closer examination of the subject's records, it was seen that although the baseline IQs at 10-11 years of age were no different between users and non-users, by age 12-13 they were showing a consistent and regular deterioration, up to 4 IQ points less.
This was observed to have begun already before cannabis use
was initiated, and was matched by their twins.
No twin studies appear to have been done yet on the association between cannabis and psychosis. However a couple of family and sibling comparisons have been done so far which have shown similar results for schizophrenia that the twin studies had for IQ.
|A Controlled Family Study of Cannabis Users with and without Psychosis||
The controlled family study done in new York is a familial comparison of whether cannabis may or may not cause schizophrenia in adolescents.
The study compared the risk of psychosis in the families of 4 different groups of people: one group non-psychotic who had never used drugs; another non-psychotic who had used cannabis heavily and no other drugs; a third group diagnosed with schizophrenia with no drug use; and the final group schizophrenic who had used cannabis heavily and no other drugs.
In comparing the familial risk of psychosis, they determined that the family members of the schizophrenia patients had a greater risk of psychosis, regardless of whether they used cannabis or not themselves. This suggested that cannabis had not caused the schizophrenia, but there was a pre-existing familial likelihood of psychosis.
|Association Between Cannabis Use and Psychosis-Related Outcomes Using Sibling Pair Analysis in a Cohort of Young Adults||
Another study was done on non-twin sibling pairs in Australia.
Or rather, the study was done first on 3801 young adults in general, and then some analyses re-done on 228 sibling pairs to confirm the results. In its introduction, the authors express the concern that no sibling pair studies of the subject have been done so far, and that if a significant association between cannabis use and psychosis is not detected in sibling pairs it could "Seriously Weaken the Argument" for cannabis causing psychosis.
Unfortunately this study was made more than usually difficult to get meaningful results out of simply because of the measure given for the time of onset of cannabis use, coupled with the fact that the subjects varied in age at testing from 18-23. The measure of cannabis use is "How many years ago did the subject first use cannabis", answers were 3 or less/4-5/6 or more. Of course, two subjects who both instigated cannabis at age 15 would give very different answers to that question if one was currently 18 and one was 23.
Added to that is the point of disorders like schizophrenia mostly becoming apparent around age 20 onwards, and you have a problem. Most of the 18 year olds may not have yet shown many symptoms, whereas most of the 23 year olds would have, and the 23 year olds would tend to have 5 years more cannabis use under their belt just because they are older.
Therefore immediately the subjects with longer cannabis use could tend to have experienced more psychotic symptoms. These factors have to be adjusted for in the analysis before any results could be reached. They did adjust for the increased likelihood of psychosis in relation to age (and gender) in all analyses, so this factor can be considered dealt with as best it can be in this study design. Unfortunately every time an adjustment has to be made more potential for inaccuracy is introduced, so it is a shame that the study measurements were taken the way they were. A better study design at the outset (such as testing all subjects at the same age) would have been nice.
On the subject of age vs cannabis use however there is a potential for confounding remaining unaddressed. They did not show the average "years since first use of cannabis" figures across the age groups. Logic would tend to indicate that the age at onset would average more or less the same for the older as the younger subjects, and therefore those at 23 would average 5 years more cannabis use than the 18 year olds. The study authors claim that because the subjects were assessed within a "relatively narrow" age range, longer duration since first cannabis use is "equivalent" to an earlier age of first use.
This generality is OK as the average population of the study should be around the middle of those age groups, 20-21 years old. However I do note that the 'range' of age of first cannabis use measures is 3 years, and the 'range' of age overall is 5 years, making for quite a lot of potential variation within this general parameter. Therefore writing it off as "equivalent" may not be accurate in all cases and a better action at this point would have been to use the figures of "time since first use" and "current age" to produce the figure of "Age at first use" on a case-by-case basis, then run the analysis on this figure instead.
Bearing in mind those limitations, the results from the initial analysis of all 3801 subjects showed that those with psychotic symptoms on average initiated cannabis use 1 year earlier. These results tended to reduce when confounding factors like hallucinations at 14 years were adjusted for, and depression and anxiety at 21.
Next they analysed the sibling pairs. There were only 10 sibling pairs where one sibling was psychotic and one was not (and no sibling pairs where both were psychotic, so there were in fact only 10 psychotic individuals in the sibling pairs, making for quite a small sample group after all).
In these they discovered that although the affected siblings showed a 1.5 year earlier onset of cannabis use, this difference disappeared when adjusted for age and sex. Remember none of these were twins, so one sibling was necessarily at least 9 months older than the other, probably more, making him or her more likely to have developed psychotic symptoms and more likely to have used cannabis longer.
The analysis of the remaining 218 non-psychotic diagnosed sibling pairs as a whole, and individually of the 100 pairs where both used cannabis, showed a small but significant tendency for the sibling with the earliest use of cannabis to have a higher score than their sibling on a rating system of delusions. 1 extra point for each year of earlier commencement was the average change. It is important to note that although the general point of the subject being around the same age was OK for evaluation of the group overall where the vast majority were at the average age, in the case of the sibling pairs there should be a greater difference in age than the average of the rest of the group.
As we already covered, the difference in age was adjusted for as regards the incidence of psychosis. However the age as regards cannabis intiation remains unaccounted for under the assumption that longer duration of use is "Equivalent" to earlier onset of use. This may not be as accurate for the sibling pairs as for the unrelated group as a whole.
Adolescent exposure to alcohol, tobacco, and other drugs is not controlled for in this study either. Other confounding factors such as socio-economic status are satisfactorily controlled in the design phase of the study by the inclusion of sibling pairs in the analyses. However the likelihood of the siblings who used more cannabis to have also used more of these other substances remains unaccounted for, and may be contributing to the effect seen.
The studies all more or less run out at this point and end with a Well we still need to check one more thing ... sort of note.
So what have we actually got out of the data available?