2. Cannabis and Its Effects
THE REVIEW OF CANNABIS EFFECTS: SOME GENERAL CONSIDERATIONS
The Scope of the Review
This review of cannabis effects deals primarily with more recent human studies, although considerable attention was directed to the earlier literature and to animal studies in preparing this summary. In many instances, these latter reports are of questionable value here, both with respect to scientific validity and social relevance. On the other hand, it is interesting that very few significant findings of cannabis effects have emerged from the more recent controlled pharmacological studies which were not at least suggested in the earlier literature. Apparently, considerable effort has gone into "separating the wheat from the chaff" in previous methodologically limited reports.
In general. references to animal studies will be restricted to the few areas where significant contributions have been made in the direction of establishing or suggesting effects in humans. The majority of animal experiments on cannabis effects have used enormous doses and questionable techniques of administration, and provide little data which can be generalized to other species, especially man. In addition, most animal behavioural studies have been concerned with effects of undetermined relationship to the major areas of interest with respect to human use. While this may be less true of biochemical and physiological studies, even here considerable inter-species differences often drastically restrict generalizations. For example, the rodent cardiovascular response, the rabbit eye blink reflex and dog ataxia indices frequently noted in the literature are. to a certain extent, idiosyncratic to those species and provide limited information regarding others.
While animal toxicology studies are de rigueur in classical medical pharmacology, many investigators have questioned the wisdom of focussing limited time and financial resources at this time on cannabis studies involving animal species and tests of questionable applicability to man. The assumption is often made in toxicity studies that one can accurately estimate the effects of long-term use of moderate amounts of a drug on the basis of information obtained from sub-chronic administration of massive doses to lower species. The predictive validity of such procedures has not been clearly established and is the subject of much controversy. Admittedly, in some situations, one has little choice but to experiment on animals. In certain circumstances, such studies have led to significant advances in human pharmacology, but in most behavioural areas this has been the exception rather than the rule. Additional effort should be made to emperically develop animal tests which adequately parallel effects and conditions of primary interest in humans. We will make occasional references to animal experiments, especially those involving primates, where they are deemed appropriate and relevant.
It may be important to note that the vast majority of both human and animal pharmacological studies of cannabis have employed only male subjects. With very few exceptions, when general information about the effects of the drug is sought, males are used females are rarely studied scientifically except when some specifically female characteristic is under consideration. To be sure, there are often justifiable biological, social and practical reasons for excluding female subjects in certain studies, but the information gap which now exists in this area may well be significant.
In human studies, two basic approaches have emerged. Some experiments were clearly designed to test for effects of the drug on some specific behaviour or function of independent interest, such as automobile driving or visual acuity, for example. The second approach is more open-ended and concerned with exploring and delineating the general or dominant characteristics of the drug response. In the area of psychopharmacology, until recently, few tests had been developed or refined for sensitive assessment of the major characteristics of the cannabis 'high'. Many standard instruments are inappropriate and insensitive to the dominant effects or may merely quantify the response to the drug on some often arbitrary and abstract dimension of little social relevance.
The Question of Dose
With all drugs, responses differ, both in the intensity and character of the reaction, according to the amount of the drug administered. The relationship between the dose and the intensity of an effect is often referred to as the doseresponse or dose-effect relationship. For every drug there is a dose low enough so as to produce no noticeable reaction, and at the opposite extreme, some degree of toxicity or poisoning can be produced by any substance if enough is taken. No drug can be designated either safe, beneficial, or harmful without consideration of the dose likely to be consumed. Consequently, it is usually essential to study a drug's effect over a range of doses in order to obtain an adequate understanding of the nature of the response. It is, of course. important to consider doses which have some relevance to existing or poten . tial patterns of' use, if social implications are to be inferred from experimental findings.
Accurate quantitative information regarding the potency of the materials used in even recent cannabis experiments is questionable due to technical problems in chemical analysis, but it would appear that the majority of the existing experiments have been more concerned with finding a dose which produces an effect, rather than with exploring effects at doses and under conditions of primary social concern and relevance.
The general question as to the appropriate range of cannabis doses to study is made more complicated by evidence of considerable individual differences in sensitivity to the drug, even when subjects have had a similar past-use history, and a growing body of information regarding the importance of the set and setting in determining effects. The situation is further compounded by evidence that chronic frequent users may develop some degree of tolerance and may use acute doses which are likely to cause untoward effects in novices or more casual users. In addition, significant behavioural adaptation to some of the initially disrupting or distracting aspects of the 'high' appears to occur with even limited experience with the drug, and such effects cannot be explained by "classical tolerance". In addition, there is frequent mention of a "reverse tolerance" or increased sensitivity to the drug after initial use in some individuals. (Factors related to tolerance are discussed in detail in a later section.) Consequently, in some circumstances, the doses studied may have to be adjusted to the individual characteristics and previous drug experiences of the subjects involved.
Until recently, relatively little was known about the actual quantity of psychoactive material administered and absorbed by users. Recent advances in chemistry have enabled some progress in this area, and more sophisticated social research has also contributed significantly.
It would appear that there is little experimental research which provides information on cannabis doses typically used in North America today. While high-dose acute studies are of course necessary and of interest in themselves, since a certain minority of regular users undoubtedly consume relatively large quantities of cannabis, a significant gap apparently exists in our knowledge of the effects of cannabis at doses and patterns of use which most frequently occur in North America and other industrial countries. While there is evidence that certain extreme chronic users of hashish in the East, for example, may smoke several hundred milligrams of THC per day 45,414,444 (often in water pipes) and might be able to tolerate acute doses approaching one hundred milligrams, there is growing evidence from a variety of sources that in North America, most users smoke less than ten milligrams of THC to get 'high' or 'stoned'. 8,111,276,296,362,397,413,414,431,436,661
A number of studies indicate that regular users typically smoke anything from a fraction of a 'joint' up to a few cigarettes in a session (which may extend over several hours). 68,89,248,276,321,413,415,475 The size of marijuana cigarettes may range from a few hundred milligrams up to a several gram 'bomber', but numerous investigators have reported that a typical 'joint' is usually less than a gram.92, 276, 305, 362, 363, 411, 413 Commission research confirms these estimates.[a] Although the potency of various samples of marijuana and hashish varies over a vast range, the notion that 'good average weed' is approximately 1% THC in North America is commonly expressed in the literature and has some documented support (although existing Canadian data suggest that about half that figure might be a more accurate estimate for
this country-see Table 2). 280, 305, 362, 397, 414, 478, 503, 66l, 678 If 1% THC marijuana were about average for experienced users, then a 'typical' 400 mg 'joint' would contain 4 mg THC. Since users often report that one or two 'joints' per person is a typical dose for a full evening session, it would appear that a single acute dose of five to ten milligrams in most individuals would produce substantial effects equal to or greater than those usually sought. Commission socio-pharmacological research and laboratory experiments provide considerable support for this notion.
In addition to survey, interview and participant observation investigations, the Commission conducted an exploratory study of characteristics of cannabis use in a primary group of 14 employed persons or students who had developed a rather stable pattern of consumption, with
most smoking several times a week. 238 These subjects were asked to keep a daily diary of drug use over a period of one month, using standard information forms assessing such variables as identity and quantity of drugs used, other persons involved, events associated with drug use and a subjective rating as to the relative intensity of the 'high' achieved. Specific weights of marijuana or hashish consumed were obtained with standard scales in many instances, and the samples involved were quantitatively analysed for cannabinoid content. (In this study the median THC content [k] for marijuana and hashish was approximately 1% and 5%,
Although complicated problems arise in the analysis of the data obtained, some general estimates of THC consumption can be made. We have no proof of the accuracy of the self-reports or of the quantities of THC actually consumed. One hundred sessions (averaging 2.28 individuals each) provided data for the following figures. Cannabis was most often taken in small groups of two or three persons, although isolated use and large groups were also described. Estimates of the mean THC used per individual in each session ranged from 1.3 mg to over 50 mg ~9 THC, in several instances, with a median of 6.2 ma. THC consumption tended to be slightly lower with marijuana than with hashish. It should be stressed that sessions were
typically several hours long (but ranged up to 12 hours), and that these figures do not generally represent a single acute dose. These estimates are, at best, a conglomerate average based on marijuana and/or hashish (occasionally used together in a session) in either cigarettes or pipes,
and is further complicated by minor alcohol consumption in approximately one-third of the sessions. The figures obtained, however, generally corroborate the estimates of "typical acute doses" obtained from other sources.
In Commission laboratory experiments we found that when smoking techniques were used which maximized the delivery and absorption of THC, cigarettes containing 5 to 10 mg THC, completely smoked, produced effects which were beyond the range of cannabis experiences of some regular users who had been 'turning on' several times a week for a number of years. As
will be discussed in more detail later, acute adverse anxiety reactions occurred in a few individuals at these doses. Subjects indicated that cigarettes containing about 6 mg THC produced effects generally comparable to those typically experienced when 'high' or 'stoned'. Furthermore, Miles and associates reported that when cannabis users were allowed to select
their own marijuana doses daily for a period of several weeks in the laboratory, on the majority of occasions, subjects consumed 2-4 mg of THC to get 'high'.43 While considerable limitations must be placed on generalizations among various laboratory and social conditions of use, the
estimates of typical acute doses obtained from a number of independent sources are in general agreement.
The Importance of Time
Another important pharmacological concept is the time-response relationship or the relation between the time which has elapsed since drug administration and the effects produced. Such a tcmporal analysis may be restricted to immediate or short-term (acute) effects of a single dose, or on the other extreme, may involve the long-term effects of persistently repeated (chronic) use of a drug. Studies of shorter periods of repeated administration are often referred to as sub-chronic.
The intensity, and often the character or quality of a drug's effects may change substantially within a short period of time after administration. An initially stimulating effect, for example, may later change to one of sedation. With some drugs, an initial state of tension or anxiety may later turn into one of relaxation and a sense of well being, or vice versa, as a function of time. Consequently, it is often essential to obtain measures at several points in time.
It is also important to consider the long-term consequences of chronic drug use (especially at higher doses). Often such effects cannot be readily predicted from what is known of the immediate response. For instance, while there is little doubt that the smoking of a few tobacco cigarettes has no lasting detrimental effect on lung or cardiac function, there is increasing
scientific evidence that long-term heavy use of this substance has serious consequences. As another example, the clinical picture of the chronic alcoholic involves psychological and physiological disturbances which do not develop with intermittent drinking of shorter duration. In simple terms, it is essential to ask "How much?", "How often?", and "For how long?" (as well as "By whom?") when discussing the long-term reaction to repeated drug use.
Commission Experiments on Cannabis Effects
The Commission has undertaken four experimental projects concerned with the acute effects of cannabis and, in some instances, of alcohol, in humans. The experiments covered a wide range of measures, but were primarily focussed on obtaining time- and dose-response effects in certain crucial areas. One experiment was concerned with the effects of cannabis and alcohol on certain automobile driving tasks. A second study was designed to determine some of the effects of cannabis and alcohol, both alone and in combination, on psychomotor tracking performance. A third study examined the effects of cannabis on visual signal detection (attention and vigilance) and. secondarily. the recovery of dim-light visual acuity after glare. The largest experiment was designed to assess the effects of cannabis on a variety of' subjective. behavioural and physiological measures, and to quantitatively and qualitatively compare synthetic high purity Delta 9 THC with natural marijuana of identical Delta 9 THC content on these variables. In all experiments subjects were paid volunteers, mostly university students, who had experience with alcohol and cannabis but no history of heavy use of either these or other drugs. Subjects were psychiatrically screened and those with detectable pathology excluded.
Standard .4 gm cigarettes of marijuana were used in all studies except one, in which both THC and marijuana were smoked. Cannabis doses were administered on an individual body weight basis under standardized conditions of smoking which maximized THC delivery and absorption. THC quantities studied in the main experiments ranged from 9 mcg/kg to 88 mcg/ kg, which, in a typical subject, resulted in minimum and maximum doses of 0.67 and 6.6 mg THC. The lowest dose was used in only one study, to determine and compare the threshold for detectable reaction for pure THC and for marijuana. The higher dose was intended to approximate the effects typically sought by regular, confirmed cannabis users in North America. Analysis of the data from these experiments indicates that although considerable individual differences exist among subjects, we were generally successful in approximating these limits. Intermediate doses were also studied. Apparently, very little experimental information as to cannabis effects in this dosage range is currently available, although differences in administration techniques and cannabis THC content analyses among various studies make definite comparisons of the actual doses absorbed by the subjects difficult. Three of our studies had two or more drug conditions in common, thereby facilitating certain comparisons among the experiments, and providing a broader data base for generalizations from a number of common measures than has usually been available in the past.
References to the findings of these various experiments are made throughout the general review which follows, and more detailed summaries are presented in Annex A at the end of this chapter. General interpretation and discussion of the findings are presented in the main text, however. The Commission's experimental program will be described in considerably more detail in supplementary technical reports. In addition to these specific experiments, new information as to the effects of cannabis has also been obtained from a variety of Commission sources: social, survey and interview studies of cannabis users in Canada; exploratory investigations of the experience of the medical profession and of others working in treatment facilities; special Commission symposia; and numerous briefs presented to the Commission.
Pharmacologically Active Constituents of Cannabis
It is widely accepted that one isomer of THC ( Delta 9) is the major active compound in hashish and marijuana. Mechoulam424 and others have argued that it is the only significant active constituent and can be generally considered pharmacologically equivalent to crude cannabis. This contention is supported by a variety of chemical and pharmacological evidence.
In 1967, Isbell and associates, using 'early' samples of relatively pure Delta 9 THC administered orally or by smoking, reported that experienced cannabis users readily recognized the subjective effects of the compounds as "marijuana-like".296,298 In addition, the physiological reactions observed in these experiments were similar to those usually attributed to crude cannabis. Although the quantity and purity of the THC used by Isbell and associates is questionable in light of what is now known about the isolation, synthesis and quantitative analysis of cannabinoids, these general qualitative observations have been corroborated on numerous occasions.655 Hollister's group, for example, has investigated the effects of Delta 9 THC on humans over the past several years, 280 and these studies have confirmed earlier observations as to the qualitative similarity between THC and marijuana, on both subjective and objective measures.
Until recently, the only other naturally occurring compound which had been shown to have significant behavioural effects was Delta 8 THC. While this isomer is undoubtedly psychotropic in humans, it is generally present in only minute amounts, if at all, in most marijuana and hashish samples. Cannabidiol (CBD) and cannabinol (CBN), the only other typically prevalent cannabinoids, have not been shown to have any substantial behavioural activity when administered orally in humans298 or intravenously in monkeys.424 The only direct pharmacological comparison that has been done between Delta 9 THC of high purity, and crude cannabis preparations has been done with monkeys.424 These studies showed a general parallel between certain behavioural and physiological effects of Delta 9 THC and a petroleum ether extract of hashish at various matched doses administered intravenously. No other lipid-soluble hashish fractions tested in these animals showed any evidence of significant activity nor was there any evidence of interaction among cannabinoids.
Recently, a number of researchers have challenged the notion that a single isomer of THC can account for the pharmacological action of cannabis in humans and animals. 111,161,222,223,280,307,309,430,655,679 In some instances such objections have been raised because of frequent accounts that experienced users report some minor subjective response to extracted marijuana placebo in laboratory experiments, in spite of the confirmed absence of cannabinoids in the material. It has also been noted that experienced cannabis users often claim the ability to differentiate among varieties of cannabis and contend that various samples may produce noticeably different effects which are, to some extent, independent of dose.248,438 Furthermore, there have recently been several papers reporting some pharmacological activity in previously unidentified cannabinoids and, in some instances, non-cannabinoid substances 49,225,341,429 Gill and associates have suggested that certain peripheral atropine-like effects of cannabis may be accounted for by water-soluble noncannabinoid compounds recently identified and tested in animals.225
Isbell and associates, in the original studies, were not able to directly compare, in a controlled quantitative fashion, all of the various cannabinoid compounds investigated, but tentatively concluded that the "pure" THC compounds tested had considerably less pharmacological activity than cruder THC and marijuana materials. The researchers also found some suggestion of THC and CBD interaction in humans.296,298 Quantitative conclusions from this early work are restrained by the comparatively primitive cannabis chemistry of five years ago, and by the limited range of effects assessed. Recently there have been suggestions of CBD and CBN activity at high doses in some animal species. 128,372,491,662 In addition, several problems emerge in the interpretation of some aspects of the monkey studies of Mechoularn and associates.424 Although there appear to be considerable objective behavioural and physiological similarities between monkey and human reactions to cannabis, no adequate techniques have been employed to measure the subjective aspects of the drug response in these animals. Since in humans cannabis effects are most striking in this phenomenological area, a significant gap remains regarding the effects which reportedly reward or reinforce use in society. Furthermore, since lipid-soluble extracts of hashish were studied (intravenously), it is possible that minor water-soluble compounds may have been missed.
Current data suggest that burning has relatively uniform effects on the relative proportions (non-acid) of cannabinoids and does not substantially introduce new compounds, but the pharmacological equivalence of cannabis administered intravenously, orally or by smoke inhalation has not been established. This fact further limits the generality of both Mechoularn and Isbell's work. At least one metabolite of THC has been shown to be pharmacologically active, and since both metabolites and unchanged THC tend to become rapidly protein or albumin bound in plasma, there may be a possible interaction or competition among cannabinoids for metabolic enzyme or binding sites. Recent evidence that CBD and CBN may alter the metabolism of THC and other drugs is also relevant.346,491,662 Consequently, even if Delta 9 THC were the only psychotropically active constituent, other cannabinoids might alter the general cannabis response indirectly through interaction with THC.
In summary, it is clear from the literature that Delta 9 THC produces subjective and physiological effects which are similar in many respects to those elicited by marijuana and hashish, but no direct quantitative comparison has previously been done in humans of the effects of relatively pure isolated or synthetic THC and crude cannabis materials or other cannabinoids. In many scientific circles, there has been an almost complete and often unquestioned reliance on the Delta 9 THC content of cannabis as a basis for standardizing and comparing different studies, even though the pharmacological equivalence among different cannabis forms accomplished with this procedure had not been directly tested. One of the Commission projects was designed to provide an initial step in bridging this gap.
The first phase of the project was to compare Delta 9 THC with natural marijuana under standard conditions.[p] Three doses of synthetic high purity Delta 9 THC in hexane-extracted alfalfa, three doses of marijuana of matched Delta 9 THC content, and an extracted alfalfa placebo were given under controlled standardized conditions to a group of 14 experienced male cannabis users. The lowest dose was aimed at establishing general pharmacological threshold values and the upper dose was intended to approximate the 'high' usually sought by regular users in North America.
Based on a review of the literature and preliminary laboratory work, a condensed test battery was constructed which was intended to assess most of the major acute cannabis effects previously identified experimentally, and to tap some of the prominent subjective features of the cannabis 'high' as reported by users. This battery was begun immediately after smoking and was repeated four hours after drug administration. Certain isolated measures were taken more frequently, and a few tests were given only once. Measures employed in this experiment included subjective ratings of the intensity of the 'high' and post-session quantitative and qualitative descriptions, a specially designed multiple choice self-report questionnaire, the Clyde Mood Scale135, heart pulse rate, salivation, conjunctival congestion, finger temperature, tonic skin conductance, visual imagery, auto-kinetic movement, two-flash fusion threshold, visual spiral after-effect, time production and time estimation, finger painting, speech performance, momentary and sustained muscle strength, tapping speed, short-term memory and other cognitive functions. In addition, experimenters rated each subject's behaviour. At the end of each session subjects were encouraged to compare the nature of the 'high' obtained on the various days.
Of 21 variables measured, 13 were shown in this experiment to be generally sensitive to cannabis effects and seven of these exhibited orderly dose- and time-response functions. No consistent qualitative or quantitative differences were found between equivalent doses of the crude marijuana and the Delta 9 THC on the various measures described. Although the interpretation of a few of the specific measures is still somewhat ambiguous, we have concluded that the two cannabis materials tested are not significantly different. This finding does not exclude the possibility that differential effects might have been obtained on other measures, or that different strains of marijuana or hashish might give slightly different results. Since our marijuana sample was unusually low in other cannabinoids (more than 90% Of the total cannabinoids were Delta 9 THC), generalizations from this experiment must be restricted accordingly. The direct investigation of Possible interaction effects among smoked THC, CBN and CBD, in various doses and combinations, had been anticipated in follow-up, but time limitations have prevented further study. Until further data are available, it would seem prudent to specify CBN and CBD, as well as THC,content of materials used. Carboxylic acid fractions should also be specified if the drug Is to be given orally.
It should be stressed that this study was focussed on the acute effects of cannabis and can provide only suggestive evidence as to the pharmacological equivalence of the chronic administration of THC and crude cannabis preparations. It is possible that other cannabis constituents might contribute to chronic effects that were not detected in this short-term study.
It has been suggested in the literature that because of possible psychotropic effects of extracted marijuana placebo preparations, some other plant material which resembled cannabis in taste and appearance might better be used as placebo.430 In one Commission experiment, extracted alfalfa was used as placebo and in the other three. hexane-extracted marijuana served this control function. (This extraction does not remove water-soluble compounds.) The placebo marijuana was prepared from a different cannabis strain than the active material used in the experiments. Cigarettes were smoked under identical conditions in all four studies, using similar subjects, and no evidence of a differential response to these two placebo materials was obtained. While some measureable effect of smoking placebo material occurs, these reactions are probably a general response to smoke inhalation (including carbon monoxide) and the inhalation techniques involved, as well as to factors associated with the users' past experiences and expectations.
Quantitative Comparisons Among Cannabis Studies
In spite of attempts to increase the precision of cannabis studies by specifying the quantities and THC content of the material involved, even recent investigations must be interpreted generally and detailed comparisons are often difficult to make. Specific quantitative comparisons among different experimental and social studies are confounded by a number of methodological and technical difficulties and inconsistencies, among them:
(1) As previously discussed, all of the earlier and most of the present quantitative analytic methods in use for estimating cannabinoid content in cannabis give generally unreliable and erratic results, both within and between laboratories. The margin for error in this regard is often considerable.
(2) While certain aspects of cannabinoid degradation and shelf life have recently been elucidated, storage conditions of samples used in many studies were apparently not optimal for potency stability (in most reports this information is omitted), and the actual potency of the materials at the time of use may have been considerably different from the original assay.
(3) Techniques of cannabis administration have not been standardized. The differences in response to cannabis given orally, by injection or through smoke inhalation have not been fully explored. It is known that substantial differences exist with various smoking techniques. Studies published in the last two years have included the use of regular pipes. water pipes and various sizes of cigarettes. Although conclusive data are yet unavailable, it is likely that these different smoking techniques result in different degrees of THC ultimately delivered to the smoker due to such factors as differential side stream loss, pyrolysis and condensation ol'volatile cannabinoids before the smoke is inhaled. In addition, in many experiments the total cigarette is not consumed or this information is not presented. Since the 'butt' or 'roach' may contain more than 20% of the THC in the original cigarette, this can be a significant factor.618
(4) Even with a consistent system of THC delivery, different patterns of' smoke inhalation and breath retention result in different quantities of THC absorbed. The standardization of oral doses might ultimately be simpler, but in North America today it would be considerably less relevant. Although absorption of cannabinoids from the gastrointestinal tract is generally complete, results obtained from oral administration are often erratic, and may be significantly influenced by the vehicle employed, and a number of unanswered questions exist as to the effects of stomach food content and gastric enzyme and acid conditions on the rate and extent of the absorption and metabolism of cannabinoids. Although intravenous injections of small quantities of synthetic THC in alcohol solution have apparently been successfully used, this technique has obvious limitations. The water insolubility of natural cannabinoids prevents simple injection techniques.
Standard smoking techniques, maximizing delivery and absorption of specified doses, would seem optimal at the present time, even if such rigidly controlled smoking regimes result in some sacrifice of ecological fidelity. Alternatively, some researchers feel that a certain subjective state (for example, a 'typical high') should be the criterion for dose, and quantities given to different individuals should be adjusted accordingly.
Recently McGlothlin, in a series of excellent reviews of the cannabis literature, has gone to considerable effort to specify doses administered in each study discussed and even to estimate THC content of materials not originally specified by the investigator.411,414 While such detail is clearly of value in some respects, we are of the opinion that for the purposes of this report, focussing on specific quantitative aspects of most studies is generally not justified for the previously stated reasons, and may merely provide an illusion of precision (and often contradiction) where none exists. Consequently, in this review, with the exception of Commission investigations (and the annexes to this chapter), specific doses used in the various studies discussed are not presented, although general comments are often made about the cannabis quantities involved.