(
(
GUEST EDITORIALS
The Medical Use of Marijuana: The Case for Clinical Trials Richard Doblin, MPP Mark A. R. Kleiman, PhD
,"
Whether or not smoked marijuana has important medical uses is a hotly contested issue. Schwartz .and Beveridge assert that their 1990 survey of 180 clinical oncologists demonstrates that smoked marijuana, even if legally available, "would not be used much more frequently by American oncologists ... 1 Schwartz and Voth make the same claim about the results of their 1994 survey of about 1500 Richard Doblin is a PhD candidate and Mark A. R. Kleiman is Associate Professor of Public Policy, John R Kennedy School of Government, Harvard University, 79 John F. Kennedy Street, Cambridge, MA 02138. Address correspondence to: Richard Doblin, 1801 Tippah Avenue, Charlotte, NC28205. Journal of Addictive Diseases. Vol. 14(1) 1995 © 1995 by The flaworth Press, Inc. All rights reserved.
5
6
JOURNAL OF ADDICTIVE DISEASES
clinical oncologists. They contrast their findings with those of a 1990 survey we conducted of 2,430 members of the American Society of Clinical Oncology (ASCO) which found that "substantial numbers of patients are not getting the medical care their doctors would prefer to provide to them because Federal law prohibits marijuana's medical use. "2-4 . In the end, the issue of the medical use of smoked marijuana must be decided by controlled clinical studies rather than surveys; opinions, even clinical opinions, are no substitute for data. Yet surveys can help scientists decide whether clinical trials are -worth the time and effort to conduct. We think that the results of our survey and those of Schwartz and Beveridge, and Schwartz and Voth, make a compelling case for conducting clinical trials with smoked marijuana. .'
REPORTED NEED FOR MEDICAL MARIJUANA - BY CLINICAL ONCOLOGISTS In the survey by Schwartz and Beveridge, 27% of respondents would prescribe marijuana more frequently to at least a few patients if it were legally available (6% much more frequently and 21% slightly more frequently). In the survey by Schwartz and Voth, 32% would prescribe mariju-ana to their patients if it were legally available (23% of these would do so fewer than 11 times a year, 5% 11-25 times, and 4% more than 25 times).5 In our survey, 48% of respondents would prescribe marijuana to at least a few patients if it were legally available (4% to many patients, 44% to a few), while another 30% reported that they needed more information before deciding if they would prescribe it. 2 While marijuana might not be the most appropriate medicine for a majority, or even a very large minority, of chemotherapy patients, some oncologists believe that there is a subgroup for whom it would be the medicine of choice, either alone or in combination with other antiemetics with other mechanisms of action. Indeed, Schwartz and Beveridge report that 23% oftheir respondents with 10 years or less of practice, and 11 % of respondents with more than 10 years of practice, affirm that there is an irreplaceable need for smoked marijuana or THC (unfortunately, this question did not distinguish the
two drugs nor did SchwaI1 of respondents who thoul these drugs). Schwartz anc gists currently recommem smoked marijuana, potenti Another intriguing fmdin~ that smoked marijuana aJ more frequently the great« mild to moderate nausea patients with severe nause;
CORRE In support of their argul marijuana would not affec appear both to misinterpre of a question we asked. T differences in the perceivi smoke (45% of responden synthetic, orally administe tive." Contrary to SchwaJ significant preference for The percentages they quot which we asked responden marijuana and oral THC. felt that marijuana was me marijuana and oral THC \ that oral THe was more question asking if oral TH directly if smoked marijua dents with opinions (and 6 Schwartz and Beveridg of the respondents who ret the subject, in the Doblin juana should be more wid ally, 53% of the responde! should be accepted as leg one-third that Schwartz ru
'J[SEASES
findings with those of a ·s of the American el hiCtrlound that " substanle medical care their doclUse Federal law prohibits use of smoked marijuana udies rather than surveys; ) substitute for data. Yet ~r clinical trials are ·worth lk that the results of our :ridge, and Schwartz and ucting clinical trials with
~AL
MARIJUANA .OGISTS dge, 27% of respondents y to at least a few patients lore frequently and 21 % ,Schwartz and Voth, 32% ts if it were legally availhan 11 times a year, 5 % ).5 In our survey, 48% of ar 5t a few patients if it ntS;-44% to a few), while more information before appropriate medicine for )f chemotherapy patients, )group for whom it would In combination with other on. Indeed, Schwartz and lents with 10 years or less h more than 10 years of lIe need for smoked mariIn did not distinguish the
Guest Editorials
7
two drugs nor did Schwartz and Beveridge report the total number of respondents who thought there was an irreplaceable need for these drugs). Schwartz and Voth report that 5% of clinical oncologists currently recommend or tolerate their patients' illegal use of smoked marijuana, potentially making them accessories to a crime. Another intriguing rmding reported by Schwartz and Beveridge is that smoked marijuana and oral THe capsules were prescribed 'more frequently the greater the degree of nausea. In patients with mild to moderate nausea, marijuana or THC ranked ninth: In patients with severe nausea, marijuana or THe ranked sixth.
CORRECTION OF ERRORS In support of their argument that the legal availability of smoked marijuana would not affect many patients. Schwartz and Beveridge appear both to misinterpret our findings and scramble the meaning of a question we asked. They report that we found "no significant differences in the perceived antiemetic effectiveness of marijuana smoke (45% of respondents believed it to be effective) and that of synthetic, orally administered THe which 42% believed to be effective. tt Contrary to Schwartz and Beveridge's claim, we reported a significant preference for smoked marijuana among oncologists. The percentages they quote actually come from a single question in which we asked respondents to compare the effectiveness of smoked marijuana and oral THe. Of the respondents with opinions. 45% felt that marijuana was more effective than oral THC, 42% felt that marijuana and oral THC were equally effective, and only 13% felt that oral THC was more effective. 6 Our survey did not include a question asking if oral THC was or was not effective. When asked directly if smoked marijuana was effective, fully 89% of the respondents with opinions (and 63% of all respondents) said it was. Schwartz and Beveridge also erroneously claim that "one-third of the respondents who returned the questionnaire, with opinions on the subject, in the Doblin and Kleiman study believed that marijuana should be more widely accepted as ·a legal medicine. tt Actually, 53% of the respondents with opinions believed that marijuana should be accepted as legal medicine. The much lower number of one-third that S~hwartz and Beveridge cite refers to the situation,
8
JOURNAL OF ADDICTIVE DISEASES
clearly described in our paper, when all respondents are included in the analysis, even the 38% with no opinion, in effect treating "no opinion" as equivalent to "no." Schwartz and Beveridge mistakenly remark that our survey "was mailed out to six times more respondents" than theirs. We actually mailed surveys to 13.5 times more oncologists, and had 7.3 times more respondents. As did they, we included only US respondents and reported survey results only for those respondents who were clinical oncologists.
ISSUES OF DEBATE Schwartz and Beveridge make a point of claiming that their higher response rate (78% vs. 43%) makes their survey more accurate than ours.7 The difference in response rate was partly spurious, and partly reflected methodological choices they made that may have purchased response rate at the price of added bias. Since some of our questions asked about sensitive anq potentially illegal acts (the use of smoked marijuana is against the law although a medical necessity defense can be asserted), we felt it best to conduct a completely anonymous survey with only one mailing. 8 On the other hand, Schwartz and Beveridge had an assistant note the postmarks of their respondents and then send out up to two additional mailings to'those oncologists who did not respond to the previous mailings. 9 This certainly must have suggested to the recipients of the survey that their responses were being noted, possibly influencing their responses. In addition, Schwartz and Beveridge's response rate was artificially inflated through a simple technique. Sample members wh~ did not reply after three mailings were replaced by substituting the next eligible person from the membership list of the American Society of Clinical Oncology.9 This essential methodological information was absent from their report. Schwartz and Beveridge err when they report that "only anecdotal reports and the Doblin and Kleiman survey claim superiority of the crude drug over synthetic dronabinol." A pilot clinical trial conducted by Vinciguerra et al. found that 29% of patients who failed to obtain relief from oral THC were helped by smoked marijuana, and concluded "Our results demonstrated that inhalation
marijuana is an effective 1 vomiting due to cancer che of references. An official J ture about the results of concluded "the data accu gram's operation [with ~ smoked results in a highe ingested. "II Researchers } (CBD), one of the main plant, reduces the anxiety the side effect profile of ~ that of the oral THC cap~ THC absorbed through th( able than that of THC abso Schwartz and Beveridg THC and smoked marijuan cial cost." While the cost e same as oral THC, the cos would be only a small fract smaller fraction of the cost Schwartz and Beveridge well as ours, was conduct( ondansetron (Zofran), whi< icine that makes all altemal Schwartz and Beveridge d contravention of the pub! Glaxo Pharmaceuticals, thl study.14 The enthusiasm f( and Beveridge paper, and universally shared. Ondans judgement that it rep res en existing therapies (not inch cal trials used to obtain FD that it is not the last war clinical trials, only 65% of fewer emetic episodes. Ful emetic episodes. IS Thus , cantly changed the pictur·
Guest Editorials
9
DISEASES
~spondents are included in
ic
1 effect treating "no J mark that our survey "was ;" than theirs. We actually logists, and had 7.3 times Ided only US respondents Ise respondents who were
4TE int of claiming that their ~s their survey more accue rate was partly spurious, lices they made that may of added bias. Since some ll1d potentially illegal acts he' law although a medical felt it best to conduct a one mailing. 8 On the other sistant note the postmarks to two additional mailings to the previous mailings.9 Ie recipients of the survey pc ')ly influencing their eittrg"e's response rate was :hnique. Sample members re replaced by substituting ,rship list of the American essential methodological :y report that "only anec11 survey claim superiority nol." A pilot clinical trial :hat 29% of patients who e helped by smoked mari10nstrated that inhalation
marijuana is an effective therapy for the treatment of nausea and vomiting due to cancer chemotherapy. ,,10 This paper was in our list of references. An official report to the New Mexico State Legislature about the results of a state-sponsored FDA-approved study concluded "the data accumulated over all five years of the program's operation [with 250 patients] do show that marijuana smoked results in a higher percentage of success than does THe ingested. "II Researchers have also demonstrated that cannabidiol (CBD), one of the main cannabinoids present in. the marijuana plant, reduces the anxiety provoked by THC.12 ThIs suggests that the side effect profile of smoked marijuana may be preferable to that of the oral THC capsule. Furthermore, the bioavailability of THC absorbed through the lungs has been shown to be more reliable than that of THC absorbed through the gastrointestinal tract. 13 Schwartz and Beveridge make the misleading claim that oral THC and smoked marijuana are "available at about the same flOancial cost." While the cost of black market marijuana is roughly the same as oral THe, the cost of legally available smoked marijuana would be only a small fraction of the cost of oral THe (and an even smaller fraction of the cost of ondansetron). Schwartz and Beveridge highlight the fact that their survey, as well as ours, was conducted before the prescription availability of ondansetron (Zofran), which they suggest is an important new medicine that makes all alternative antiemetics. ess necessary. Although Schwartz and Beveridge did not choose to mention it (in apparent contravention of the published editorial policy of this journal), Glaxo Pharmaceuticals, the maker of ondansetron, supported their study.I4 The enthusiasm for ondansetron reflected in the Schwartz and Beveridge paper, and in Schwartz and Voth's editorial, is not universally shared. Ondansetron's IB rating from the FDA reflects a judgement that it represents only a modest therapeutic gain over existing therapies (not including smoked marijuana). Even the clinical trials used to obtain FDA-approval for ondansetron demonstrate that it is not the last word in antiemetics. In one of the largest clinical trials, only 65% of the recipients of ondansetron had two or fewer emetic episodes. Fully 21 % of its recipients had five or more emetic episodes. IS Thus we doubt that ondansetron has significantly changed the picture painted by our respondents, 21 % of
10
JOURNAL OF ADDICfIVE DISEASES
whom found the lack of efficacy of currently available antiemetics to be a problem in half or more of their patients, and 54% of whom found that antiemetic side effects were a problem for more than a few patients. In the face of these numbers and the results of the ondansetron trials, refusal to pursue the clinical potential of whole cannabis seems to us to reflect an unjustified complacency about the adequacy of the current antiemetic armamentarium. Schwartz and Beveridge raise the issue of the deleterious effect of marijuana on driving. While this is an important concern, a recent study sponsored by the National Highway Traffic Safety Administration concluded that marijuana "produces a moderate degree of driving impairment ... its magnitude is not exceptional in comparison to changes produced by many medicinal drugs and alcohol. Drivers under tlJ,e influence of marijuana retain insight in their perfonnance and will compensate where they can, for example, by slowing down or increasing effprt. As a consequence, THe 's adverse effects on driving perfonnance appear relatively small. "16 These fmdings hardly distinguish marijuana from other . drugs routinely given to cancer patients. Schwartz and Voth raise the issue of possible negative interactions between marijuana·and other hepatically-metabolized medicines, and also claim that it is not only possible but probable that the medical use of marijuana would impair some patients' short-term memory and result in their forgetting to take other life-saving medicines. There is . little reason to suspect marijuana would be substantially different in these risks than the oral THe capsule, for which no such negative outcomes have been reported in the literature in the more than eight years that the oral THe capsule has been legally prescribed.
QUESTIONABLE CONCLUSIONS Schwartz and Beveridge report that their survey "included independent analysis by a consulting biostatistician which reduced statisticalbias." Unfortunately, a careful reading of their three conclusions suggests that statistical bias should not have been their only concern. They conclude that "(1) marijuana, the crude drug which is smoked, or orally administered synthetic THe may be helpful but are certainly not indispensable for selected adult oncology patients
with post-chemotherapy rl tainly true for some selectc and 23 % of their own res} placeable need for these dr Their second conclusi< "(2) cannabis derivatives adverse effects." Whether not cannot be detennined f data on the rate of adversl point of comparison, in om with headaches in 23% of ents, abdominal pain in 49. adverse effects is problem idge survey does not distin . effects, asking only if the Further complicating this f that bothersome adverse e 110 [respondents] who re scribed synthetic THe as ~ the respondents experience patients. If this interpretati the 'total proportion of all I could be 25% is if every ! who reported adverse eff. effects. This is extremely u meant to report that advers of 25% of the patients of I Beveridge actually found i since they have failed to Sl even though we have requf Schwartz and Voth ass( bothersome side effects are who have used marijuana I plausible, clinical oncolog fact into their clinical judg intend to suggest that rna! for their prior illicit drug I most appropriate treatmenl
')[SEASES
:ntly available antiemetics at; ' <;, and 54% of whom , p.....Aem for more than a ers and the results of the ;linical potential of whole :tified complacency about namentarium. e of the deleterious effect an important concern, a l Highway Traffic Safety la "produces a moderate litude is not exceptional in any medicinal drugs and narijuana retain insight in vhere they can, for examffort. As a consequence, Irmance appear relatively !ish marijuana from other ssible negative interactions rtetabolized medicines, and obable that the medical use ;' short-term memory and saving medicines. There is e substantially different in Ir which no such negative w ~ the more than eight .gauy prescribed. ~LUSI0NS
:ir survey "included inde;tician which reduced stading of their three conclunot have been their only ma, the crude drug which c THe may be helpful but :d adult oncology patients
Guest Editorials
IJ
with post-chemotherapy nausea or vomiting." While this is certainly true for some selected adult oncology patients, between 11 % and 23% of their own respondents reported that there was an irreplaceable need for these drugs for some patients. Their second conclusion, repeated by Schwartz and Voth, is "(2) cannabis derivatives have a high rate (25%) of bo~hersome adverse effects." Whether this figure represents a "high rate" or not cannot be detennined from the survey since they did not collect data on the rate of adverse effects for any other antiemetics. As a point of comparison, in one clinical trial ondansetron was associated with headaches in 23% of recipients, constipation in 5% of recipients, abdominal pain in 4%.17 The interpretation of the 25% rate of adverse effects is problematical because the Schwartz and Beveridge survey does not distinguish between minor and serious adverse effects, asking only if there were "bothersome" adverse effects. Further complicating this fmding, Schwartz and Beveridge reported that bothersome adverse effects "were experienced by 25% of the 110 [respondents] who recommended smoked marijuana or prescribed synthetic THe as antiemetic drugs." It follows that 75% of the respondents experienced no bothersome adverse effects in their patients. If this interpretation is correct, the only possible way that the total proportion of all patients who experienced adverse effects could be 25% is if every single patient of the 25% of respondents who reported adverse effects actually experienced these adverse effects. This is extremely unlikely. Perhaps Schwartz and Beveridge meant to report that adverse effects were experienced by an average ' of 25% of the patients of the 110 respondents. What Schwartz and Beveridge actually found is difficult for us to determine, especially since they have failed to supply us with a copy of their survey form even though we have requested it for more than three years. Schwartz and Voth assert that the patients least likely to suffer bothersome side effects are younger, marijuana-experienced patients . who have used marijuana nonmedically. If this is the case, as seems plausible, clinical oncologists could be expected to incorporate this fact into their clinical judgement. Surely, Schwartz and Voth do not intend to suggest that marijuana-experienced patients be punished for their prior illicit drug use by being denied what is for them the most appropriate .treatment.
12
JOURNAL OF ADDICI'lVE DISEASES
Schwartz and Beveridge's third and fmal conclusion is that "natural marijuana would not be used more frequently by 76% of oncologists, even if legal restrictions on its medical use were liberalized. " Despite the "independent analysis by a consulting biostatistician," Schwartz and Beveridge slightly overstate their case. They previously reported that 27% of their respondents would prescribe marijuana more frequently if legal restrictions were liberalized. This results in a total of only 73%, not 76%, who would not do so. More importantly, Schwartz and Beveridge ignore the methodological problems with their survey that would tend to bias their results downward. Furthermore, they seem to imply that the patients of the 27% of respondents who would prescribe marijuana if it were legally available (32% according to the Schwartz and Voth survey) do not deserve to receiv~f the medicine that their oncologists would prefer to prescribe. Moreover, they ignore the possibility that some physicians who would not prescribe whole marijuana based on currently available knowledge might be influenced by the results of clinical trials. Our survey found that about a third of oncologists were waiting for more data before making up their minds, a practice that deserves to be more widely observed.
FROM SURVEYS TO CLINICAL TRIALS Survey research has contributed about as much as it can to this debate. All three surveys report that a substantial percentage of oncologists would prescribe marijuana if it were legally available. Clinical trials should therefore proceed to test the clinical utility of marijuana as an antiemetic in cancer chemotherapy. Anecdotal evidence convinces us that clinical trials should also be conducted into the ·use of smoked marijuana as an appetite enhancer in the treatment of the HIV-related wasting syndrome, to reduce the pain of spasticity, and for other indications as well. Though Schwartz and Beveridge appropriately point out the health risks of illicit marijuana when it is contaminated with fungus and aspergillus, simple sterilization methods could be used if marijuana were to be legally available. The development of smoke filtration systems (water pipes) or non-combustive means of volatilizing the active agents in crude cannabis (vaporizers), in order to address concerns about the
lung insult from marijuar (Doblin) is actively engag both the wasting syndromt that Schwartz, Be~eridge efforts of researchers to sec trials with marijuana so tt basis for regulatory and cli It is unfortunate that tl entangled in the debate ( non-medical use of cann: purely medical decision, t cannabis would have no policy. 18,19 The fact that c( medical use) rather than suggests that the public, a: tion between medical and ponents of drug legalizati< unable to do so. The fact t in cannabis, is already leE there are no published rep illicit market, further weal material must be resisted 0 Thus we hope that fut' clinical studies rather than internal medicine, oncolol utility for marijuana, rathe of drug abuse. The interesl ate treatment deserve to 0 opposing camps in the war
I 1. Schwartz R, Beveridge R it today? Opinions from clinical 2. Doblin R, Kleiman MAR oncologists' experiences and aU 3. Doblin R, Kleiman MA 1991; 114:809-810.
Gllest Editorials
DISEASES
lal conclusion is that "nat~O · 'tly by 76% of oncolIk se were liberalized. " ;onsulting biostatistician," ate their case. They pre)ondents would prescribe :rictions were liberalized. 6%, who would not do so. ~e ignore the methodologid tend to bias their results Iply that the patients of the ribe marijuana if it were Ichwartz and Voth survey) lat their oncologists would e the possibility that some 'hole marijuana based on nfluenced by the results of out a third of oncologists ~ up their minds, a practice I.
"leAL TRIALS t as much as it can to this
substantial percentage of f i' ~re legally available. ) tc.,./the clinical utility of motherapy. Anecdotal eviuld also be conducted into :tite enhancer in the treatme, to reduce the pain of ell. Though Schwartz and ealth risks of illicit marius and aspergillus, simple uijuana were to be legally filtration systems (water ilizing the active agents in iddress concerns about the
/3
lung insult from marijuana smoke, is also called for. One of us (Doblin) is actively engaged in organizing marijuana research into both the wasting syndrome and smoke filtration systems. We hope that Schwartz, Be,:,eridge and Voth will join us in supporting the efforts of researchers to secure governmental pennission for clinical trials with marijuana so that data can replace mere opinion as the . basis for regulatory and clinical decision-making. It is unfortunate that the medical marijuana issue has become entangled in the debate over the appropriate legal status of the non-medical use of cannabis. We believe that this should be a purely medical decision, because the clinical availability of whole cannabis would have no direct impact on drug abuse control policy. 18.19 The fact that cocaine is a Schedule II drug (available for medical use) rather than a Schedule. I drug (entirely prohibited) suggests that the public, as well as officials, can grasp the distinction between medical and nonmedical use. Only a few fervent proponents of drug legalization and a few fervent drug warriors seem unable to do so. The fact that THe, the primary intoxicating agent in cannabis, is already legally available for medical use and that . ' there are no pubhshed reports of leakage of medical THC into the illicit market, further weakens the case that use of the whole plant material must be resisted on drug-abuse-control grounds. .1?us we .hope that future papers on this topic will report on chmcal studies rather than surveys, and be published in journals on internal medicine, oncology, and other possible fields of medical utility for marijuana, rather than in journals devoted to the problem of drug abuse. The interests of the seriously ill in getting appropriate treatment deserve to outweigh ideological concerns of the two opposing camps in the war about the War on Drugs.
REFERENCES 1. Schwartz R, Beveridge R. Marijuana as an antiemetic drug: How useful is it today? Op.inions fr~m clinical oncologists. 1 Addic Dis. 1994; 13 (1):53-65. 2. Doblln R, Kleiman MAR. Marijuana as antiemetic medicine: A survey of oncologists' experiences and attitudes. J Clin Oncol. 1991; 9:1314-1319. 3. Doblin R, Kleiman MAR. Medical use of marijuana. Annals Int Med.
1991; 114:809-810. ,
14
JOURNAL OF ADDICfIVE DISEASES
4. Doblin R. Marijuana: A question of currently accepted medical use, in Cancer Treatment and Marijuana Therapy, ed. Robert C. Randall. Washington, DC: Galen Press, 1990:Appendix F: 341-352. 5. Schwartz RH, Voth EA. Marijuana as medicine: Making a silk purse out of a sow's ear. J Addic Dis. 1995; 14(1):15-21. 6. This finding comes from a preliminary report (ref. #4), when the total number of respondents was 1027. Our final paper (ref. #2) reported on a total of 1035 respondents. In that paper, the percentage of respondents with an opinion who felt that marijuana was more effective than oral THC changed from 45% to 44% (N 121), while the number who fell that marijuana and oral THC were equally effective changed from 42% to 43% (N = 120). The number who felt that oral THC was more effective remained 13% (N = 36). 7. Schwartz and Beveridge report that our response rate was 42%. That rate was from our preliminary report (ref. #14). In our final paper (ref. #2), we reported a 43% response rate. Schwartz and Voth claim that our response rate for completed questioMaires was actually 40%, not 42% as we claimed, and that our survey did not have a high enough response rate to be statistically valid. Their statement about our response rate claims is simply incorrect. Their statement about validity seems to confuse validity with variability. 8. It was due to the potentially illegal nature of the medical use of marijuana that we did not ask respondents about their current use of marijuana, as Schwartz and Beveridge suggested we should have done. 9. Personal telephone communication, Richard Schwartz, May, 1994. 10. Vinciguerra V, Moore T, BreMen E. Inhalation marijuana as an antiemetic for cancer chemotherapy. NY State J Med. 1988: 88:525-527. 11. The Lynn Pierson Therapeutic Research Program, Annual Report to the New Mexico Legislature, 1984. 12. Zuardi AW, Shirakawa E, Finkelfarb E, Kamiol IG. Action of cannabidiol on the anxiety and other effects produced by A9-THC in normal subjects. Psychopharm. 1982; 76:2A5-250. 13. Ohlsson A, Lindgren JE, Wahlen A, et al. Plasma .1.-9-THC concentrations and clinical effects after oral and intravenous administration and smoking. Clin Pharmacol Ther. 1980; 28:409-416. 14. This fact was reported in a description of Schwartz's survey in Substance Abuse Report, March 15, 1991:5. By contrast, the sponsorship of our study by ACT, about which Schwartz and Voth make such a fuss, was entirely nominal. The research received no outside funding. We have seen no evidence of the "multimillion dollar campaign" described by Schwartz and Voth. 15. Physicians Desk Reference, 1994: 796. 16. Hindrik WI, O'Hanlon J. Marijuana and Actual Driving Performance. Report #DOT HS 808078. November, 1993. Abstract. 17. Physicians Desk Reference, 1994: 798. 18. Kleiman MAR. Marijuana: Costs of Abuse, Costs of Control. New York: Greenwood Press, 1989. p. 167. 19. Kleber H. US drug laws-all introduction. New Eng I Med. 1994; 330 (5), Feb 3, 1994:356-365.
Avec Homm8 Con Complirr
Mit Freundlici Yoroshiku
~~ ......
S Nailuchshil C HaHJJYIIUlJ. With Complin of the authon
=
Notes: -
The Journal in w refereed professic copies are availat: 10 Alice Street, Bi
I' :
.
tn'
"
