II. THE PROBABLE HEALTH EFFECTS OF CANNABIS USE
Acute Psychological and Health Consequences
The acute toxicity of cannabis is very low. There are no confirmed cases of human deaths from cannabis poisoning in the world medical literature. Animal studies indicate that the dose of THC required to produce 50% mortality in rodents is extremely high by comparison with other commonly used pharmaceutical and recreational drugs. The lethal dose also increases as one moves up the phylogenetic tree, suggesting by extrapolation that the lethal increases as one moves up the phylogenetic tree, suggesting by extrapolation that the lethal dose in humans could not be very easily achieved by smoking or ingesting the drug (Grinspoon and Bakalar, 1993; Rosencrantz, 1983).
The most common unpleasant acute psychological effects of cannabis use are anxiety, sometimes producing unpleasant depressive feelings (Tart, 1970; Weil, 1970). These effects are most often reported by naive users who are unfamiliar with the drug's effects, and by patients who have been given oral THC for therapeutic purposes. More experienced users may occasionally report these effects after receiving a much larger than intended dose of THC. These effects can usually be prevented by adequate preparation of users about the type of effects they may experience, or they can be managed by reassurance and support (Weil, 1970).
Motor vehicle accidents
The major potential health risks from the acute use of cannabis arise from its effects on cognitive and psychomotor performance. Intoxication produces dose-related impairments in a wide range of cognitive and behavioural functions that are relevant to a skilled performance like driving an automobile or operating machinery. These include: slowed reaction time and information processing, impaired perceptual-motor coordination and motor performance, impaired short term memory, signal detection, tracking behaviour and slowed time perception. (Chait and Pierri, 1992).
The negative effects of cannabis on the performance of psychomotor tasks is almost always related to dose (Chait and Pierri, 1992). The effects are generally larger, more consistent and persistent in different tasks which involve sustained attention. The acute effects of "recreational" doses of cannabis on driving performance in laboratory simulators and over standardised driving courses are similar to those of doses of alcohol that achieve BACs between 0.07% and 0.10% (Hansteen, Miller, Lonero, Reid and Jones, 1976; Peck et al, 1986; Smiley, 1986).
While cannabis impairs performance in laboratory and simulated driving settings, it is difficult to relate the magnitude of these impairments to an increased risk of being involved in motor vehicle accidents. Students of the effects of cannabis on on-road driving performance have found at most modest impairments (e.g. Sutton, 1983). Cannabis intoxicated persons drive more slowly perhaps because they are more aware of their level of psychomotor impairment than alcohol intoxicated drinkers who generally drive at faster speeds (Smiley, 1986).
No controlled epidemiological studies have established that cannabis users are at increased risk of motor vehicle accidents. A prospective Swedish study of mortality over 15 years among military conscripts found an increased risk of premature mortality among men who had smoked cannabis 50 or more times by age 18. Violent deaths were the major contributor to this excess, of which 26% were motor vehicle and 7% other accidents (e.g. drownings and falls). The increased risk disappeared, however, after multivariate statistical adjustment for confounding variables such as alcohol and other drug use (Andreasson and Allebeck, 1990).
Uncertainty about the role of cannabis in motor vehicle accidents is likely to remain since case-control studies are difficult to conduct and interpret (see chapter by Smiley this volume). Blood levels of cannabinoids do not indicate whether a driver or pedestrian was intoxicated with cannabis at the time of an accident, and many drivers with cannabinoids in their blood were also intoxicated with alcohol at the time of the accident (Smiley, 1986). Factors other than psychomotor performance also contribute to the danger of drug use when driving. Foremost among these is the user's preparedness to take risks when intoxicated, which the available evidence suggests is reduced by cannabis intoxication by contrast with alcohol intoxication which consistently increases risk-taking (Smiley, 1986)
The Health Effects of Chronic Cannabis Use
The Immune System
There is reasonably consistent evidence that THC can produce cellular changes such as alterations in cell metabolism, and DNA synthesis, in vitro (Bloch, 1983). There is even stronger evidence that cannabis smoke is mutagenic in vitro, and in vitro, and hence, that it is potentially carcinogenic for the same reasons as tobacco smoke (Leuchtenberger, 1983).
There is reasonably consistent evidence that cannabinoids impair both the cell-mediated and humoral immune systems in rodents (Munson and Fehr, 1983). These changes have decreased resistance to infection by bacteria and viruses. There is also evidence that the noncannabinoid components of cannabis smoke impair the functioning of alveolar macrophages, the first line of the body's defence system in the lungs (Munson and Fehr, 1983). The relevance of these findings to human health is uncertain: the doses required to produce these effects are often very high, and the problem of extrapolating from the effects of these doses to those used by humans is complicated by the possibility that tolerance develops to the effects on the immune system (Hollister, 1992).
Moreover, the limited experimental and clinical evidence on immune effects in humans is mixed, with a small number of early studies that have suggested adverse effects not being replicated by later ones (Munson and Fehr, 1983; Hollister, 1992). At present, there is no conclusive evidence that consumption of cannabinoids predisposes humans to immune dysfunction, as measured by reduced numbers or impaired functioning of T-lymphocytes, B-lymphocytes or macrophages, or reduced immunoglobulin levels.
The clinical and biological significance of these possible immunological impairments in chronic cannabis uses is uncertain. To date there has been no epidemiological evidence of increased rates of disease among chronic heavy cannabis users. There is one large prospective study of HIV-positive homosexual men which indicates that continued cannabis use did not increase the risk of progression to AIDS (Kaslow et al, 1989). Given the duration of large scale cannabis use by young adults in Western societies, the absence of any epidemics of infectious disease makes it unlikely that cannabis smoking produces major impairments in the immune system.
It is more difficult to exclude the possibility that chronic heavy cannabis use produces minor impairments in immunity. Such effects would produce small increases in the rate of occurrence of common bacterial and viral illnesses among chronic users which would be of public health significance because of the increased expenditure on health services, and the increased loss of productivity that this would entail among the young adults who are the heaviest users of cannabis. A recent epidemiological study by Polen at al (1993) which compared health service utilisation by nonsmokers and daily cannabis only smokers has provided the first suggestive evidence of an increased rate of presentation for respiratory conditions among cannabis smokers. This remains suggestive, however, because infectious and noninfectious respiratory conditions were not distinguished.
The Cardiovascular System
The conclusion reached by the Institute of medicine in 1982 still stands: the smoking of marijuana "causes changes to the heart and circulation that are characteristics of stress ...[but] there is no evidence ... that it exerts a permanently deleterious effect on the normal cardiovascular system ..." (p 72). The cardiovascular effects of cannabis may be less benign in patients with hypertension, cerebrovascular disease and coronary atherosclerosis for whom marijuana poses a threat by increasing the work of the heart (Aronow and Cassidy, 1974, 1975).
The Respiratory System
Chronic heavy cannabis smoking impairs the functioning of the large airways, and probably causes symptoms of chronic bronchitis such as coughing, sputum, and wheezing (Bloom et al, 1987; Huber et al, 1988; Tashkin et al, 1988a, 1990). Given the documented adverse effects of tobacco smoke which is qualitatively very similar in composition to cannabis smoke (Tashkin, 1993; Wu et al, 1988). There is as yet little direct evidence that the latter occurs (Huber et al, 1988) although there is evidence that chronic cannabis smoking may produce histopathological changes in lung tissues of the kind that precede the development of lung cancer (Fligiel et al, 1988).
More recently, concern about respiratory cancers has been heightened by a series of case reports of cancers of the aerodigestive tract in young adults with a history of heavy cannabis use (e.g. Caplan and Brigham, 1989; Donald, 1991; Taylor, 1988). Although these reports fall short of providing convincing evidence because they were uncontrolled and many of the cases concurrently used alcohol and tobacco, they are clearly cause for concern since such cancers are rare in adults under the age of 60, even among those who smoke tobacco and drink alcohol (Tashkin, 1993). Smoking cannabis may also pose an acute risk to individuals with respiratory diseases such as asthma. Evidence linking tobacco smoke to asthma and asthmatic symptoms is increasing.
The potential adverse effects of cannabis on the respiratory system are specific to smoking as the route of administration, and could not result from oral ingestion.
Chronic cannabis use probably disrupts the male and female reproductive systems in animals, reducing the secretion of testosterone, and sperm production, motility, and viability in males, and disrupting the ovulatory cycle in females (Bloch, 1983; Institute of Medicine, 1982). It is uncertain whether it is these effects in humans, given the inconsistency in the limited literature on human males (Mendelson and Mello, 1984), and the lack of research in the case of human females (Hollister, 1986). There is also uncertainty about the clinical significance of these effects in normal healthy young adults. They may be of greater concern among young adolescents, and among males whose fertility has been impaired for other reasons.
Cannabis smoking during pregnancy probably impairs foetal development Gibson et al, 1983; Hatch and Bracken, 1986; Tennes et al, 1985; Zuckerman et al, 1989, leading to a reduction in birthweight (Abel, 1985). This may be a consequence of a shorter gestation period, and probably occurs by the same mechanism as cigarette smoking, namely, foetal hypoxia. There is uncertainty about whether cannabis use during pregnancy produces a small increase in the risk of birth defects as a result of exposure of the foetus in utero. There is some animal evidence of such effects although these studies have usually involved very high doses by the oral route (Abel, 1985). The limited studies in humans have generally but not consistently produced null results (Gibson et al, 1983; Hatch and Bracken, 1986; Hingson et al, 1982; Zuckerman et al, 1989).
There is not a great deal of evidence that cannabis use can produce chromosomal or genetic abnormalities in either parent which could be transmitted to offspring. Such animal and in vitro evidence as exists suggests that the mutagenic capacities of cannabis smoke are greater than those of THC, and are probably of greater relevance to the risk of users developing cancer than to the transmission of genetic defects to children (Bloch, 1983; Hollister, 1986).
There is suggestive evidence that infants exposed in utero to cannabis may experience transient behavioural and developmental effects during the first few months after birth (e.g. Fried 1985, 1989). There are several case-control studies which suggest that there is an increased risk of certain childhood cancers (namely, astrocytomas and leukemia) among children born to women who reported that they had used cannabis during their pregnancies (Kuitjen et al, 1990; Robison et al, 1989).
Possible Health Effects of Contaminants in Cannabis
Because cannabis is an illegal drug its cultivation, harvesting and distribution are not subject to quality control mechanisms to ensure the reliability and safety of the product used by consumers. It is well recognised in developing countries, such as Kenya, that illicit alcohol production can result in the contamination with toxic by-products or adulterants that can kill or seriously affect the health of users. The same may be true of illicit drugs such as opiates, cocaine and amphetamine in developed societies. There is no evidence of comparable health effects for cannabis although there were concerns expressed about the possible health effects of the use of cannabis contaminated by herbicides (such as paraquat) that were used to control illicit cannabis cultivation in the US in the 1970s. These concerns proved unfounded (Hollister, 1986). There have also been concerns about the microbial contamination of cannabis leaf but there has been little evidence (other than a small number of case histories) that this has adversely affected the health of cannabis users (Hollister, 1986).
Psychological Effects of Chronic Cannabis Use
One of the major concerns about the psychological effects of chronic heavy cannabis use has been that it impairs adult motivation. The evidence for an "amotivational syndrome" among adults consists largely of case histories and observational reports (e.g. Kolansky and Moore, 1971; Millman and Sbriglio, 1986). The small number of controlled field and laboratory studies have not found compelling evidence for such a syndrome (Dornbush, 1974; Negrete, 1983; Hollister, 1986). The evidential value of the field studies is limited by their small sample sizes, and the limited sociodemographic characteristics of their samples, while the evidential value of the laboratory studies is limited by the short periods of drug use, the youthful good health of the volunteers, and minimal demands made on volunteers in the laboratory (Cohen, 1982). Some regular cannabis users report a loss of ambition and impaired school and occupational performance as adverse effects of their use (e.g. Hendin et al, 1987) and that some ex-cannabis users give impaired occupational performance as a reason for stopping (Jones, 1984). Nonetheless, it is doubtful that cannabis use produces a well defined amotivational syndrome. It may be more parsimonious to regard the symptoms of impaired motivation as symptoms of chronic cannabis intoxication rather than inventing a new psychiatric syndrome.
In the United States in the 1970s and 1980s, cannabis use appears to have increased the risk of discontinuing a high school education, and of experiencing job instability in young adulthood (Newcombe and Bentler, 1988). The apparent strength of these relationships in cross-sectional studies (e.g. Kandel, 1984) has been exaggerated because those adolescents who are most likely to use cannabis have lower academic aspirations and poorer high school performance prior to using cannabis than their peers who do not (Newcombe and Bentler, 1988). It remains possible that factors other than the marijuana use account for apparent causal relations. To the extent they may exist, these adverse effects of cannabis and other drug use upon development over and above the effect of pre-existing nonconformity may cascade throughout young adult life, affecting choice of occupation, level of income, choice of mate, and the quality of life of the user and his or her children.
A major finding of research into the adult consequences of adolescent cannabis use has been the strong evidence of a regular sequence of initiation into the use of illicit drugs among American adolescents in the 1970s in which cannabis use preceded involvement with "harder" drugs such as stimulants and opioids (Kandel et al, 1984; Donovan and Jessor, 1983; Yamaguchi and Kandel, 1984 a, b). The causal significance of this sequence of initiation into drug use remains controversial. The hypothesis that it represents a direct effect of cannabis use upon the use of the later drugs in the sequence is the least compelling. There is better support for two other hypotheses which are not mutually exclusive: that there is a selective recruitment into cannabis use of nonconforming adolescents who have a propensity to use other illicit drugs; and that once recruited to cannabis use, the social interaction with other drug using peers, and exposure to other drugs when purchasing cannabis on the black-market, increases the opportunity to use other illicit drugs (Baumrind, 1983; Goode, 1974; Kandel, 1988).
A Dependence Syndrome
A cannabis dependence syndrome as defined in DSM-IV (American Psychiatric Association, 1994) can occur in heavy, chronic users of cannabis. There is good experimental evidence that chronic heavy cannabis users can develop tolerance to its subjective and cardiovascular effects. There is also suggestive evidence that some users may experience a withdrawal syndrome on the abrupt cessation of cannabis use, although one that is much milder and less marked than that experienced when withdrawing from alcohol or opiates (Compton, Dewey and Martin, 1990; Jones and Benowitz, 1976). DSM-IV notes that "symptoms of possible cannabis withdrawal (e.g. irritable or anxious mood accompanied by physical changes such tremor, perspiration, nausea and sleep disturbances) have been described in association with the use of very high doses, but their clinical significance is uncertain." (American Psychiatric Association, 1994:215).
There is clinical and epidemiological evidence that some heavy cannabis users experience problems in controlling their cannabis use, and continue to use the drug despite experiencing adverse personal consequences of use (Jones, 1984; Roffman et al, 1988; Weller et al, 1984). There is limited clinical evidence for a cannabis dependence syndrome analogous to the alcohol dependence (Kosten et al, 1987). Epidemiological surveys of the prevalence of drug dependence in the general population (e.g. Anthony and Helzer, 1991) show that cannabis dependence, as defined in the diagnostic manuals, is among the most common forms of drug dependence in Western societies by virtue of its high prevalence of use. On the other hand, relatively few users seek treatment for cannabis dependence (American Psychiatric Association, 1994: 220-221).
The weight of the available evidence suggests that even the long term heavy use of cannabis does not produce any severe or grossly debilitating impairment of cognitive function (Carter et al, 1980; Fehr and Kalant, 1983b, Rubin and Comitas, 1975; Wert and Raulin, 1986). If it did research to date should have detected it. There is some clinical and experimental evidence, however, that the long-term use of cannabis may produce more subtle cognitive impairment in the higher cognitive functions of memory, attention and organisation, and the integration of complex information (Page et al, 1988; Solowij et al, 1991, 1992, 1993 and see chapter by Solowij in this volume). While subtle, these impairments may affect everyday functioning, particularly among individuals in occupations that require high levels of cognitive capacity. The evidence suggests that the longer the period that cannabis has been used, the more pronounced is the cognitive impairment (Solowij et al, 1992, 1993). It remains to be seen whether the impairment can be reversed by an extended period of abstinence from cannabis.
A suspicion that chronic heavy cannabis use may cause gross structural brain damage was provoked by a single poorly controlled study using an outmoded method of investigation which reported that cannabis users had enlarged cerebral ventricles (Campbell et al, 1971). This finding was widely and uncritically publicised. Since then a number of better controlled studies using more sophisticated methods of investigation have consistently failed to demonstrate evidence of structural change in the brains of heavy, long term cannabis users (e.g. Co et al, 1977; Kuehnle et al, 1977). These negative results are consistent with the evidence that any cognitive effects of chronic cannabis use are subtle, and hence unlikely to be manifest as gross structural changes in the brain.
Serious Psychiatric Disorder
There is suggestive evidence that large doses of THC can produce an acute psychosis in which confusion, amnesia, delusions, hallucinations, anxiety, agitation and hypomanic symptoms predominate. The evidence comes from laboratory studies of the effects of THC on normal volunteers and clinical observations of psychotic symptoms in heavy cannabis users which remit rapidly following abstinence (Bernardson and Gunne, 1972; Chopra and Smith, 1974; Edwards, 1976).
There is less support for the hypothesis that cannabis use can cause either an acute or a chronic functional psychosis (Thornicroft, 1990). Such possibilities are difficult to study because of the rarity of such psychoses, and the near impossibility of distinguishing them from schizophrenia and manic depressive psychoses occurring in individuals who also use cannabis (Ghodse, 1986).
There is strongly suggestive evidence from a prospective study that heavy cannabis use may precipitate schizophrenia in vulnerable individuals (Andreasson et al, 1987; Schneier and Siris, 1987; Thornicroft, 1990). This relationship is still only strongly suggestive because in the only prospective study conducted to date (Andreasson et al, 1987) the use of cannabis was not documented at the time of diagnosis, there was a possibility that cannabis use was confounded by amphetamine use, and there are doubts about whether the study could reliably distinguish between schizophrenia and acute cannabis, or other drug-induced, psychoses (Negrete, 1989; Thornicroft, 1990).
There is less support for the hypothesis that cannabis use can cause either an acute or a chronic functional psychosis (Thornicroft, 1990). Such possibilities are difficult to study because of the rarity of such psychoses, and the near impossibility of distinguishing them from schizophrenia and manic depressive psychoses occurring in individuals who also use cannabis (Ghodse, 1986). There is better evidence that cannabis use can adversely affect the course of schizophrenia in affected individuals who continue to use it (Cleghorn et al, 1991; Jablensky et al, 1991; Martinez-Arevalo et al, 1994).