Comparing Cannabis with
British Medical Journal 2003;326:942-943 (3
Britain now has 13 million tobacco smokers. This number has been steadily decreasing due to public awareness of the harm caused by tobacco smoking. At the same time the number of cannabis smokers is increasing. Between 1999 and 2001, the number of 14-15 year olds who had tried cannabis rose from 19% to 29% in boys and 18% to 25% in girls, and a Home Office document estimates that 3.2 million people in Britain smoke cannabis. 1 2 However, the harmful effects of smoking cannabis are widely known and have recently been highlighted. 3 4 Although the active ingredients of the cannabis plant differ from those of the tobacco plant, each produces about 4000 chemicals when smoked and these are largely identical. Although cannabis cigarettes are smoked less frequently than nicotine cigarettes, their mode of inhalation is very different. Compared with smoking tobacco, smoking cannabis entails a two thirds larger puff volume, a one third larger inhaled volume, a fourfold longer time holding the breath, and a fivefold increase in concentrations of carboxyhaemoglobin.5 The products of combustion from cannabis are thus retained to a much higher degree. How is this likely to translate into adverse effects on health?
We already know that regular use of cannabis is associated with an increased incidence of mental illnesses, most notably schizophrenia and depression,4 but it is also worth examining its potential to cause other illnesses, especially those of the heart and respiratory system.
At present, there is an understandable dearth of epidemiological evidence of cardiopulmonary harm from cannabis, because its use is a relatively new phenomenon and its potency is changing. The amount of the main active constituent, tetrahydrocannabinol (THC), in cannabis has increased from about 0.5% 20 years ago to nearer 5% at present in Britain, whereas "Nederweed" (the variety smoked in the Netherlands) has an average of 10-11% tetrahydrocannabinol. At the same time little study has been undertaken of any concomitant change in the content of tar. Case-control studies are difficult to perform since cannabis cigarettes do not come in standard sizes, which makes dose-response relations difficult to establish. Furthermore, most users of cannabis also smoke tobacco, which makes it difficult to dissect out individual risks. As with tobacco, there will be a latent period between the onset of smoking and the development of lung damage, cardiovascular disease, or malignant change.
Tobacco smoking is responsible for 120 000 excess deaths each year in Britain, 46 000 from cancers, 34 000 from chronic respiratory disorders, and 40 000 from diseases of the heart and circulation. However, there are indications that smoked cannabis may cause similar effects to smoking tobacco, with many of them appearing at a younger age. Smoking cannabis causes chronic bronchitis, emphysema, and other lung disorders, which were recently summarised in a review released by the British Lung Foundation.3 A striking feature of cannabis smoking is that it is associated with bullous lung disease in young people.6 Inflammatory lung changes, chronic cough, and chest infections are similar to those in cigarette smokers, but may also be commoner in younger people.7-9 Premalignant changes have been shown in the pulmonary epithelium, and there are reports of lung, tongue, and other cancers in cannabis smokers.
Tetrahydrocannabinol has cardiovascular effects, and sudden deaths have been attributed to smoking cannabis.10 Myocardial infarction is 4.2 times more likely to occur within an hour of smoking cannabis.11 However, despite these alarming facts, there is no evidence at present on whether smoking cannabis contributes to the progression of coronary artery disease, as smoking cigarettes does. More studies of the cardiovascular and pulmonary effects of cannabis are essential.
It may be argued that the extrapolation from small numbers of individual studies to potential large scale effects amounts to scaremongering. For example, one could calculate that if cigarettes cause an annual excess of 120 000 deaths among 13 million smokers, the corresponding figure for deaths among 3.2 million cannabis smokers would be 30 000, assuming equality of effect. Even if the number of deaths attributable to cannabis turned out to be a fraction of that figure, smoking cannabis would still be a major public health hazard. However, when the likely mental health burden is added to the potential for morbidity and premature death from cardiopulmonary disease, these signals cannot be ignored. A recent comment said that prevention and cessation are the two principal strategies in the battle against tobacco.12 At present, there is no battle against cannabis and no clear public health message.
Competing interests: None declared.
1. Schools Health Education Unit. Young people in 2001. Exeter , 2002.
www.sheu.org.uk/pubs/yp01/yp01.htm [accessed 18 Feb 2003].
Responses sent to the BMJ about the above article
I read with interest the editorial by Henry et al. I should like to address the fact that cannabis use is as they say associated with an increased incidence of mental illness however the casual relationship remains to be proven and to follow on in the same sentence to state 'to examine its potential to 'cause' other illnesses is imprecise and may be read that we have excepted the evidence that cannabis does indeed cause mental illness. No such casual evidence exists.
The paper by Bachs et al relating to acute cardiovascular fatalities following cannabis use is far again from conclusiveand the authours themselves were cautious to stress the , and I quote , possible nature of the association. Again the text of the Henry paper reads 'attributed' clearly there is a difference again between cause and association.
In addition one must remember that many users of cannaboids are staunch antismokers and instead prefer to cook so called 'hash cakes'I wonder how many of the schoolchildren who admitted taking cannabis had done this in preference to smoking. Indeed if users are eating rather than smoking then the putative causal relationship is many times less likely.
A recent editorial suggested that in the future as many as 30,000 deaths a year in Britain may be caused by smoking cannabis(1). But this conclusion was not based on any new scientific evidence and the arithmetic appears to be based on a series of questionable assumptions.
Cannabis smoke does contain many of the same poisonous substances that are found in tobacco smoke and cannabis smokers deposit more tar in their lungs than cigarette smokers because they inhale more deeply and tend to hold their breath(2). But to expose the lungs to the same amount of tar as an average 15 - 20 a day cigarette smoker, cannabis users would have to smoke 4-5 times a day every day of the week. In fact surveys of young cannabis users in Britain suggest that very few fall into this category ¨C a large majority are occasional ¡°weekend¡± users, and even among more frequent users few fall into the high use category of 4-5 times a day(3). It is obviously impossible to get accurate statistics on the numbers of daily cannabis users, but the figure of 3.2 million in Britain cited by the authors of the editorial is far too high. It is also difficult to get accurate scientific data on the effects of regular cannabis use on the lungs because many users mix cannabis resin with tobacco. But studies of cannabis-only smokers in California showed that they do tend to develop signs of chronic bronchitis ¨C but there is no evidence that this progresses to more severe lung diseases such as emphysema or lung cancer(4).
An important factor is that unlike cigarette smokers most cannabis smokers tend to quit when they reach their 30¡¯s. Long term surveys of cigarette smokers showed that those who quit before the age of 35 had only a very slightly increased risk of lung cancer(5). The risk of developing lung cancer depends far more on the duration of smoking than on the number of cigarettes consumed. Thus smoking 40 cigarettes a day as opposed to 20 doubles the risk of lung cancer, but smoking for 30 years as opposed to 15 years increases the risk by 20-fold. If the risks of cannabis smoking equate to those of tobacco and the majority of users give up before the age of 35 they may run little additional medical risk.
The BMJ authors also suggested that the more potent forms of cannabis that are sometimes available nowadays somehow carry an increased medical risk ¨C but one could argue exactly the opposite. THC, the active chemical ingredient in herbal cannabis, is not known to be harmful to the lungs ¨C indeed there is some scientific evidence that it may possess anti -cancer properties(6). It is also known that users when exposed to more potent forms of cannabis adjust their smoking behaviour to inhale less frequently and less deeply, while obtaining the same amount of THC (2,7). The users of potent forms of herbal cannabis may thus benefit from a reduced exposure to potentially harmful tar.
Finally, the BMJ authors added some gratuitous additional warnings about the dangerous effects of cannabis on the heart. It is true that cannabis tends to stimulate the heart and it could potentially be harmful to people who have a pre-existing heart disease, but the published scientific data has not shown this to be a serious medical problem. The two publications cited are based on very small samples and circumstantial data. In Britain virtually no cases of drug-related death due to cannabis have been reported in recent years ¨C despite our strict national system for reporting substance abuse-related deaths.
While cannabis cannot be considered to be completely harmless and it does cause adverse effects on the lungs ¨C the sort of scientific/medical scaremongering indulged in by the authors of this editorial is completely unscientific and fails to advance the public health debate about cannabis. Their arithmetic simply does not add up. Instead they help to bring science into further disrepute, and make it less likely that young people will listen seriously to any health message concerned with drugs.
Professor Les Iversen PhD FRS Department of Pharmacology University of Oxford
Henry, Oldfield, and Kon sounded the alarm about potential lung problems in cannabis smokers. 1 As these authors mention despite their concerns, large studies still show little lung damage in those who smoke cannabis and not tobacco2. Nevertheless, new information may allay the fears of some readers worried about the plant's pulmonary effects.
First, the reported increase in cannabis potency does not translate into greater risk for pulmonary problems. Though many authors argue that estimates from the 1970s of .5% THC are clearly inaccurate, most believe current data suggesting that THC concentrations average near 5% and can reach as high as 20%. The stronger cannabis, however, yields less tar per unit of THC than weaker cannabis, and leads to less deposition of tar into the lungs of smokers 3.
Because problem users are often reluctant to abstain from cannabis completely, health care professionals might suggest ways to increase the safety of the drug. The common habit of holding smoke in the lungs for extended periods provides greater exposure to noxious materials. This practice should be actively discouraged. At least 3 studies show that longer breathhold durations have little meaningful impact on intoxication 4.
In addition, vaporizing cannabis rather than smoking it can create the same subjective effects with no exposure to many toxins.5 Vaporizers have become readily available and relatively inexpensive. These machines have the potential to eliminate pulmonary problems associated with cannabis use. Smoking small amounts of potent cannabis through a vaporizer and refraining from holding smoke in the lungs presents little risk of lung troubles.
Mitchell Earleywine, associate professor of clinical science Department of Psychology, University of Southern California, Los Angeles, CA, USA 90089-1061 email@example.com
1. Henry, J. A., Oldfield, W. L. G., Kon, O. M. Comparing cannabis with tobacco, BMJ 2003; 326:942-943.
2. Polen, M. R. Health care use by frequent marijuana smokers who do not smoke tobacco. West J Med 1993; 158: 596-601.
3. Matthias, P., Tashkin, D. P., Marques-Magallanes, J.A., Wilkins, J. N. & Simmons, M.S. Effects of varying marijuana potency on deposition of tar and delta9-THC in the lung during smoking. Pharmacol Biochem Behav 1993; 58: 1145-1150.
4. Azorlosa, J. L., Greenwald, M. K., Stitzer, M. L. Marijuana smoking: effects of varying puff volume and breathhold duration. J Pharmacol Exp Ther 1995; 272: 560-569.
5. McPartland, J.M., Pruitt, P.,L. Medical marijuana and its use by the immuno-compromised. Altern Ther Health Med 1997; 3:39-45.
The brief editorial, "Comparing Cannabis with Tobacco" by Henry, Oldfield and Kon (British Medical Journal, May 3) is so full of gross inaccuracies, unsupported assumptions and unjustified speculation that, were it not for the fact that it decries chronic use of cannabis, it's unlikely a reputable journal would publish it.
For starters, there is considerable evidence that cannabis provides significant relief of a wide variety of troublesome symptoms; given fierce US prohibition policy (as reflected in United Nations treaties), data supporting that contention has been understandably difficult to gather and publish. Beyond that, the editorial clearly infers that the only way cannabs is used is by smoking in the manner the authors describe at some length. Actually, that manner of smoking, while still too popular, was fokelore based on erroneous assumptions; it has been repudiated in responsible articles for years.
There are actually many ways cannabis can be ingested other than smoking; perhaps the best is vaporization, a technique endorsed by knowledgeable clinicians because it preserves the rapid cerebral feedback that allows a user to avoid unwanted intoxication and also largely eliminates the harms of smoking.1
This brief response permits only passing reference to work I am now engaged in-- thanks to the passage of a medical cannabis initiative in California (Proposition 215). That law, passed in 1996, has effectively allowed certain chronic users to 'come in from the cold ' as it were. Systematic interviews of several hundred reveal them to be a very specific population which is surprisingly uniform in the way they became chronic users, their lifetime use of other psychotropic agents (including alcohol and tobacco), and also in their belief that sustained moderate use of cannabis has afforded them considerable benefit over years and-- in many cases-- decades
Seen in that context, the editorial is a throwback to "Reefer Madness," the mid-Thirties American propaganda campaign which eventually led to creation of a huge criminal market for one the most versatile and useful plants ever cultivated. It's unlikely that the editorial will deter that market significantly, but it certainly contributes to the confusion precluding adoption of an intelligent drug policy; one based on factual evidence rather than empty dogma.
1) D. Gieringer, "Cannabis Vaporization: A Promising Strategy for Smoke Harm Reduction," Journal of Cannabis Therapeutics Vol. 1#3-4: 153-70 (2001)
Competing interests: Member of several organizations advocating drug policy
Thomas J O'Connell MD
'Comparing cannabis with tobacco'. Henry JA, Oldfield WLG, Kon OM. BMJ (2003) 326: 942-943
I would commend these authors on using the correct scientific term cannabis, unlike some colleagues who seem to prefer terms allegedly introduced by governments rather than scientists.
That said, this is one of the most un-scientific BMJ articles I have read. Despite their being opposites in most respects, Henry and co-authors try to compare cannabis and tobacco. While both are common psychoactive drugs, cannabis is a relaxant, tobacco a partial-stimulant. One is highly addicting, the other is not. One has been prescribed by physicians down the ages and continues to be recommended in certain clinical circumstances by doctors of good repute. Hence a 'comparison' is an intriguing concept unless clearly stated objectives are being examined (eg. dependency, mortality, side effects, beneficial effects, etc). .
Cannabis has an extremely low mortality while tobacco's toll is legion. Nearly 20,000 Australians die from tobacco related disease each year with few if any cannabis reported deaths.
When examining any drug, one looks for costs and benefits but these authors have only looked for 'costs' and, for cannabis, then they can only point to 'associations'. Even if cannabis actually caused some cases of mental disease (and it does induce dependency in a small proportion of heavy users), the drug may also alleviate some conditions such as anxiety, insomnia, depression, anorexia or chronic pains.
These authors state that it might be seen as 'scaremongering' to speculate on the basis of cannabis being of equal toxicity as tobacco ... yet they go ahead and do just that: "the corresponding figure for deaths among 3.2 million cannabis smokers would be 30,000" [annually in the UK]. Can these authors be serious when no group of suspected cases is yet to be reported after the drug has been used for thousands of years in western society? If they are interested in speculation, why don't they look at alcohol consumption in cannabis smokers?
Quite apart from their tenuous position in trying to point to cardiovascular complications which may occur with smoking cannabis, they make numerous questionable and unreferenced statements in their paper including the howler about cannabis strength increasing over the years (by 10 to 20 times!). Even if this were true, it would mean less by-products for the same amount of drug and thus possibly safer smoking. Also, cannabis can be taken orally with no effect on the lungs at all, but these authors do not canvass that issue, nor other harm reduction steps. Without references, they also quote "Nederweed" ('the variety smoked in the Netherlands') which they claim has an *average* of 10-11% tetrahydrocannabinol. This is obviously unhelpful since Holland, like other countries, has a variety of cannabis and resins available on the market, including cannabis cookies.
These authors make much of the increase in cannabis use and the reductions in tobacco consumption in recent years. However, they are not open enough to discuss the legal status of the drugs. If these authors are honestly concerned about harms from cannabis then it is hard to understand why they would ignore the spectacular failing of current prohibitions in addressing these harms. The results of long term cannabis decriminalization (eg. South Australia, Holland) are equally ignored by these 'scaremongers' (to use their own term).
Andrew Byrne, Dependency Physician
Oldfield and Kon observe, quite correctly, that cannabis consumption is rising while tobacco consumption is declining in the United Kingdom. This is true in many other countries and has been the case now for several decades. These developments are no accident. In numerous countries, achieving a decline in tobacco consumption required a steadfast committment to policy based on research evidence concerning which prevention measures work and which do not. The critical public health achievment of tobacco control has been won despite the immense power of the tobacco industry. In contrast, vast resources have been allocated to law enforcement efforts to reduce cannabis smoking with very little benefit identifiable and much in the way of unintended adverse consequences. Surely if there is a lesson to be learnt from this, it is that those who are concerned to reduce the prevalence of cannabis smoking should support the same measures that worked so well for tobacco. Tobacco control has been achieved within the framework of a taxed and regulated drug. There is virtually no support among tobacco control experts for the re-introduction of tobacco prohibition. A sustained decline in cannabis consumption will only be achievable when the drug is taxed and regulated like tobacco and policy is based on evidence.
The retention of cannabis prohibition despite the lack of success and the high financial and social cost of this policy, has required a 'talking up' of the toxic effects of cannabis. Cannabis is not by any means innocuous. But the health and other adverse consequences of cannabis are dwarfed by those of alcohol and tobacco. This point was made by several reputable authors in a recent WHO review that was dropped following political pressure.
One of the many costs of cannabis prohibition is the publication in reputable medical journals of highly questionable commentary on the relattive toxicity of cannabis.
Dr Alex Wodak,
Director, Alcohol and Drug Service St. Vincent's Hospital, Darlinghurst, NSW
Sir, I read with interest the editorial of Henry et al (1). Tobacco smoking is the greatest evil in lung cancer risk across 90% of all lung cancer cases. However, evidence suggested that cannabis use is not far behind. Recent trends in lung cancer mortality in the United States showed that there is an apparent 'birth cohort' effect in lung cancer risk after 1950 (2). The authors' have speculated that this observation may be attributed to cannabis smoking, in addition to changing tobacco-smoking habits across the populations.
Interestingly, one of the cannabinoids (delta 9-tetrahydrocannabinol) has shown to have an apparent beneficial effect on lung adenocarcinoma in animal models (3). By contrast, there is accumulating evidence of histopathologic and molecular changes in lung tissue of smokers, suggesting cannabis could increase lung cancer risk in humans (4). Is this a paradoxical observation? In addition, cannabis smoke contains many of the same carcinogens found in cigarettes, as pointed out by Henry et al (1).
Until the cause-effect relation of cannabis on human health, including lung cancer, is clear, is it 'scientific' or rather 'premature' to contemplate on cannabis cessation programmes in line with tobacco smoking programmes?
1. Henry JA, Oldfield WLG, Kon OM. BMJ 2003: 326: 942-3.
Competing interests: ZK is a Research Fellow in Lung Cancer Epidemiology at
the University of Dublin (Trinity College).
To the editorial board
I was surprised at some of the things said in the recent BMJ editorial. "Can you compare cannabis with tobacco?"
In future, when someone editorialises on such a contentious issue, could you please ask them to declare their political allegiance. In particular do Dr Henry and Dr Oldfield support the current practice of criminalising cannabis smokers? Should cannabis smokers be locked up in prison? This is an issue they ignore but it is the major public policy used to discourage cannabis use. In fact, the editorial states that "At present, there is no battle against cannabis and no clear public health message." Are they unaware that unauthorised possession of cannabis is a criminal offence? Does a 'War on Drugs' not 'battle' against cannabis?
Cannabis and Tobacco can't be compared in this way.
In a ranking of addictivity of 6 drugs both Henningfield (NIDA) and Benowitz (UCSF) ranked Nicotine as the most addictive and marijuana as the least addictive (comparing Nicotine, Heroin, Cocaine, Alcohol, Caffeine and Marijuana). [Hilts, P.J. The New York Times 2-Aug-94, C3]
The writer states that "there are indications that smoked cannabis may cause similar effects to smoking tobacco, with many of them appearing at a younger age"
But cannabis smokers
While tobacco smokers:
It is impossible for me to understand how the writers arrive at their 'comparison', given that one of them is a consultant, specialist registrar at a Department of Respiratory Medicine.
I'm all in favour of improving research into the harmful effects of smoking and of discouraging the smoking of anything but scaremongering and/or criminalisation are not the way to do it.
Competing interests: None declared
If concern for our children's future is indeed your primary motive, should you not include the harm done to our children who must now somehow try to succeed in life now with the ball and chain of a criminal record with them.. "Them" is our children. The "outcast" status that a criminal record bestows on our brothers, mothers and fathers, and of course our children, and it's effect on their future needs no study to determine how destructive it is on their lives. It is severe. And this is obvious. There is no controversy here. It is common sense.
I snipped your references because you referenced nothing. You gave an opinion.....your own opinion. You stated that cannabis "impairs" the chemical transmission system. You also inserted the word "badly". Which study used those value judgements?. Interfering with the bodies own chemistry or natural functions is how drugs work. All drugs. Aspirin and caffeine for example. This interference is not inherently a bad thing, as you seem to be suggesting.
No, it's time that reality rather than blind hysteria is brought to the subject of drugs. For example; have you ever been given a shot of morphine in the hospital? Did you know that heroin is in fact nothing more than morphine that has been slightly altered so as to pass the blood:brain barrier more quickly, providing a faster onset of the morphine? . Yes... morphine. The heroin, once past the blood brain barrier reverts back to morphine and from that point on, the high is indistinguishable from morphine because it is morphine. My point is that the demonization of heroin has succeeded in turning a useful drug into something that no one of their right mind would ever want to do. Well unfortunately, millions of people, and perhaps you, although convinced that heroin will kill them and is immediately addictive, etc, have in fact, for all intents and purposes, already done it.
That is the power of misinformation Ms. Brett. Thousands of people in jail for doing a substance that doctors are giving to patients, in the hospitals, by the bathtubs full, daily.
If you want to help our children, please rethink your stance on drugs. Driving the users of the seriously harmful drugs like the stimulants, (methamphetamine and cocaine) into back-alleys only exacerbates the problem. Drugs will not go away using laws. Countries that summarily execute drug-dealers or users have not stopped it. Drugs, whether you like it or not, are here to stay. All that can be done is to reduce the damage done. Common sense.
Competing interests: None declared
Editor, The editorial by Henry et al on cannabis is quite simply the most unbalanced and inappropriate piece of writing on this subject I have seen for some time. It puts together questionable assumptions, wooly science and urban myths (such as the "potency" of modern cannabis) which conflict with the vast majority of reputable current literature. One must ask what the authors reasons were for this article - it could hardly have been to educate the profession.
And, as Wodak notes, such scare tactics are not likely to do much good - entrenching hardline prohibitionist policy will, based upon 50 years of evidence (rather than rhetoric), only increase the damage from cannabis, most of which stems from its prohibition, not the drug itself.
Ashton raises the issue of 8 year old smokers without, apparently, asking how these kids come to have the drug, where their parents or teachers are, and whether these kids may have problems apart from cannabis use which may impact upon their health and wellbeing. Easier to blame the drug, perhaps but that gets us.......where? More prohibition, more money spent on a counterproductive war on drugs, and thus not on schools, welfare, equity, justice. Even if cannabis where the cause of these kids problems, do the current policies and practices prevent these problems (clearly, no) or worsen them (probably, yes).
Canada, The Netherlands and many other jurisdictions have broken away from the mesmerised trance that chanting pro-prohibition mantras induces in many otherwise thoughtful people and institutions. Time for Britain, and Australia, to do likewise.
Competing interests: None declared
There are serious problems and misunderstandings with this editorial. I would like to list several:
1. Most of the mortality risk associated with tobacco use arises from sustained use over several decades, and the risks increase sharply as lifetime exposure accumulates. The importance of lifetime exposure was underlined in a major study of tobacco smokers in 2000 which found that: "People who stop smoking, even well into middle age, avoid most of their subsequent risk of lung cancer, and stopping before middle age avoids more than 90% of the risk attributable to tobacco." . A similar pattern should be expected for CHD and COPD - the two other major tobacco-related fatal diseases.
To the extent there is data on use of cannabis, it suggests that most users (so far) quit using it in their 30s. In the OPCS Psychiatric Morbidity Survey carried out in 1993, some 14% of adults aged 16-24 were users, but the figure dropped to 2% among those aged 35-44, and was less than 0.5% in people aged over 45. There may be cohort effects operating here, and it is possible that today's young people will have longer cannabis careers, but at present what this seems to indicate is that few people have accumulated 20 or more years of continuous use.
The very high risks due to tobacco use ultimately arise from its addictiveness, which causes many tobacco smokers to continue to smoke well after they would choose to stop. Over 70% of current users say they would like to stop, and over 80% regret ever starting: a sure indicator of addiction sustaining long term and heavy use. As cannabis has very different dependency characteristics (it is much less addictive) then its pattern of use is different most users smoke less and quit earlier.
2. Completely incompatible characterisations of the user population are used in the editorial. The figure of 13 million tobacco users is determined by those answering 'yes' to the question 'do you smoke nowadays'. In practice over 80% of these are daily users and the average consumption is just over 15 cigarettes per day per smoker. Tobacco / nicotine is an intensive drug-using syndrome for most of its users. In contrast, the Home Office figure of 3.2 million users quoted for cannabis is 'use in the last 12 months'. The figure for use in the last month (not quoted in the editorial) is 2,062,000. The Home Office does not assess how many use cannabis daily, but it will be very substantially less. Again the reason is grounded in addictiveness - the lower dependency-forming characteristics of cannabis allow for more occasional use than cigarette smoking, which generally consolidates into a powerful addiction needing constant attention by the user.
3. The point that THC concentrations have increased by a factor of ten over the last twenty years is dubious as a point of fact, but more importantly, it is completely misinterpreted. Put bluntly, a ten-fold increase in THC concentration does not mean that modern users are ten times as stoned as in the past. Users of both cannabis and nicotine control their drug exposure by varying how much smoke they inhale and retain. Higher concentrations of THC may therefore lead to LOWER smoke inhalation for a given drug exposure. This is well understood for tobacco (and the reason why 'light' cigarettes are such a fraud) but not well studied for cannabis - however it is unlikely that users do not control their intake or they would be ten times as stoned as they were 20 years ago. Ironically, the concern raised in the editorial about different puff volumes for cannabis (based on 1987 data, by the way) may actually have been alleviated by the asserted increase in THC concentration in the drugs now in use leading to lower smoke exposure as users control their dose by taking fewer and lighter puffs.
4. The derivation of the figure of 30,000 deaths is so facile it shouldn't really have been written down. At this stage, there is only limited evidence linking cannabis use to the big tobacco-related killers - cancer, CHD and COPD. While these links should be expected, the magnitude of the risk to the user (simply assumed to be equivalent to tobacco in the derivation of the 30,000 figure) will depend on a variety of factors, in particular the lifetime exposure and patterns of use - and these are very different indeed. Very few of the 120,000 smoking-related deaths occur in people under 40, yet hardly any of the users of cannabis are over 40 - so who are the 30,000 dying? Given that the smoking careers differ so much, and the usage patterns are so different, the estimate of 30,000 deaths is ridiculous. Qualifying the calculation by saying it may be 'a fraction' of that adds nothing if we don't know whether the fraction in question is one half or one-thousandth. It does leave the media-sensitive headline number in place and puts the figure into the public domain as the only estimate. It is sure to be used by those with agendas other than forming rational evidence-based insights into public health issues.
5. The case has not made that cannabis is a 'major public health hazard' as asserted in the editorial. It is certainly not harmless and the authors suggest several harmful effects. But there is a continuum between 'harmless' and 'major public health hazard' and simply showing there are dangers is insufficient to place a phenomenon like cannabis on that continuum. Most credible reviews to date have tended to suggest limited public health impacts. For example, the Advisory Council on Misuse of Drugs , concluded in March 2002 after a thorough review of the evidence... "The high use of cannabis is not associated with major health problems for the individual or society." There is always a need to challenge such assessments, but any challenge has to be credible.
6. To say there is no battle against cannabis when it is a criminal offence (even after reclassification) to use it, grow it or sell it is absurd. I agree that more could be done to promote understanding of the harm it causes and I hope the findings about the link between cannabis schizophrenia, which appear to settle the question over the direction of causation, are filtering through to users. However, one reason why health promotion efforts sometimes fail is the lack of credibility of the arguments presented to users. The casually fabricated mortality figure and 'war-on-drugs' rhetoric of the editorial are wholly counter-productive in that regard. (Incidentally, the illegal status of cannabis is a barrier to wider and better understanding of its risks because it denies opportunities for mandatory labelling and inserts in the packaging.)
Finally, the finding that cannabis is not harmless is not new and adds little to the important and highly-charged debate about its legal status, which is really about societal management of personal risk and relationship between the state and the individual. Understanding of addictiveness and its impact on personal choice and patterns of consumption are crucial in positioning different drugs, and entirely absent from the analysis presented in the editorial. Sadly, editorials like this play well in a particularly rabid section of the popular media, which has no interest in a thoughtful societal response to all drugs based on harm-reduction, respect for civil liberties and cost effectiveness. Rather than fanning the flames of tabloid ignorance, the BMJ is usually a beacon of rational and measured debate on these vital issues. I fear the editorial guard may have been down on this one.
I don't think it is a competing interest, but in the interest of clarity I would like to disclose that I was Director of Action on Smoking and Health (UK) until March 2003. I am writing in a personal capacity.
 Peto R et al. Smoking, smoking cessation, and lung cancer in the UK since 1950: combination of national statistics with two case-control studies. BMJ 2000; 321: 323-329.
 Advisory Council on the Misuse of Drugs. The classification of cannabis under the Misuse of Drugs Act 1971, (UK Government) Home Office, March 2002 (5.1).
Competing interests: None declared