This chapter looks at the bad outcomes arising from drug use as reported by students. It examines the official cases as well to see what kinds and how many instances of ill effects came to the attention of school administrators, local police, and school medical services during the study period. It also compares the rate of these official cases with the unofficial rates of illicit use as indicated by our survey findings and considers other indices or estimates of student drug experience. In this task the focus is on average estimates of illicit use on each campus by non-users and less intensive and intensive users of drugs. The chapter also looks at the students who reported bad effects and compares them with students using the same drugs who reported no ill effects.

In discussing the data, we cannot assume that all bad effects were reported; perhaps, the students suffering the worst ones could not stay in school to be included in our sample or, perhaps, they were the noncooperators on the several campuses where noncooperation occurred. It is also likely that students did not report all of their ill effects—some because they could not recall them, others because they were embarrassed, and still others because they may not have recognized an ill effect when it happened to them. Ill effects, like other forms of unpleasantness, depend upon the standards of the beholder. This is true even for medical diagnosis where unreliability can be quite high (see Blum, 1964) even for well-known disorders. In the case of drugs, where the acute effects themselves can impair judgment and recollection and where no "objective" observer is recording outcomes, the underreporting of ill effects is taken as a foregone conclusion. Consider Hollister's (1968) work, for example, in which he found subjects taking LSD reported a number of distressing physical symptoms during the acute phase. These same subjects asked about their LSD experience a few weeks later could not recall the transient ill effects. A more complex case centers about personality changes which a conventional psychologist or psychiatrist would term undesirable; a drug user experiencing these changes might disagree. This phenomenon was encountered in our earlier study (Blum and Associates, 1964). It is illustrated in the extreme by the contention of Leary and some others that a toxic psychosis or other longer-term psychosis following LSD is a good thing. Another illustration arises from arrest and incarceration which most conventional people would consider an undesirable consequence of drug use. Some of thestudents we have met have claimed otherwise, saying that jail was a good experience. The foregoing problems are all part of the larger dispute over "drug abuse" as such and cannot be resolved here. There is one virtue in the subjective standard employed here: it represents the conviction of the drug user himself that something bad has occurred. Most observers would agree that such an evaluation represents at least anxiety if not something worse and, as such, deserves attention.

It should be kept in mind that the ill effects reported here are based on lifetime experience. They do not represent any count of the frequency of such effects per person—that is, they are not incidence figures. Rather, they are prevalence figures showing simply the num • ber of persons who have experienced an ill effect one or more times in their lives. Under this system no person can report more than once any one ill effect; he can, however, report as many different ill effects as we have different codes to record them. We had 30 different codes.

BAD OUTCOMES REPORTED

To begin with, let us report that among the 75 per cent of the students who have used tobacco, only 55 per cent replied when asked whether there were any benefits, gains, or pleasures to be had from the drug. What that implies is that 45 per cent find nothing good about it after trying it. How many say there are specific bad or unpleasant effects? Forty-nine per cent, or about half the users, report unpleasantness. These are, in order of frequency for physical difficulties, headache (35 per cent of all users), other physical distress (13 per cent), and appetite difficulties (5 per cent). Among psychological and social problems, the criticism of smoking by friends and relatives ranks first (36 per cent) followed by the student's worry about his own overuse (27 per cent), troubles with friends arising from smoking or smoking incidents (8 per cent), emotional upset perceived as a tobacco effect (3 per cent), and discipline by (lower-grade) school authorities for smoking (3 per cent). We shall not list other lessoften-mentioned troubles.

Are the variety of bad effects dependent upon how much tobacco one uses, or, at least, are they reported by the kinds of people who also report heavy tobacco use? The answer is "yes." The average number of different ill effects reported by heavy smokers is one per person; for light smokers it is .5 per person.
.
First asked about benefits, 74 per cent of the student population who have used alcohol (92 per cent) say there are gains and pleasures from drinking. As for ill effects, 63 per cent of the drinkers report them. Among physical effects reported are headache and/or nausea (89 per cent) of all drinkers, slurred speech (57 per cent), appetite problems (11 per cent), unconsciousness (11 per cent), other physical effects (3 per cent), and physical injury incurred while drinking (2 per cent). Among major or formal social problems there are fights (8 per cent), school discipline for drinking (5 per cent), automobile accidents (5 per cent), moving-violation arrests (including tickets) (3 per cent), jail (3 per cent), and criminal offenses or acts while drinking (2 per cent). Among drinking problems more intimately social or psychological in nature are trouble in thinking as a consequence of drinking (21 per cent), sensitivity to the criticism of friends about one's drinking (18 per cent), loss of self-control (19 per cent), worry about one's own loss of control over use of alcohol (14 per cent), emotional upset due to drinking (11 per cent), trouble with friends (8 per cent), sexual difficulties or unfortunate sexual incidents arising from drinking (6 per cent), study problems (5 per cent), insomnia (4 per cent), and hallucinations (.08 per cent). We shall not specify less frequently occurring troubles here.

Are the variety of bad outcomes associated with the intensity of drinking? In our comparison of heavy drinkers with lighter drinkers, the average number of different ill effects reported per person for heavy drinkers is 2.3 and for light drinkers is 1.1.

Twenty-one per cent of the students have used amphetamines. Of these, 80 per cent report benefits and gains and 61 per cent report bad outcomes. Of the ill effects reported, emotional upset is the most common (35 per cent), followed by insomnia (25 per cent), headache or nausea (19 per cent), other physical distress (11 per cent), appetite problems (9 per cent), worry about one's own overuse (8 per cent), sensitivity to friends' criticism of one's amphetamine use (7 per cent), troubles in thinking when using the drug (6 per cent), loss of self-control (2 per cent), hallucinations (2 per cent), and troubles with friends arising out of incidents when using the drug (1 per cent). Are intensive users likely to report a greater variety of ill effects? Yes, for the average number of different bad outcomes reported by intensive users is 1.6 and for less intensive users is 0.8.

About one fourth of the students report they have taken sedatives. Of these, 83 per cent report benefits from use and 23 per cent report bad effects. Ill effects include worry about overuse (8 per cent), headache and/or nausea (7 per cent), other physical distress (11 per cent), sensitivity to criticism of friends or relatives about use (4 per cent), insomnia (attributed to the drug) (3 per cent), slurred speech (2 per cent), emotional upset due to the drug (2 per cent), bad dreams (2 per cent), difficulty in thinking (2 per cent), drug-induced incidents leading to trouble with friends (1 per cent), unconsciousness (1 per cent), and appetite problems (1 per cent). We find that intensive users report more diverse ill effects than less intensive users; the former average .4 ill effects per person, the latter .24.

About one fifth of the students (19 per cent) use tranquilizers. Among these 80 per cent report benefits and 17 per cent report bad outcomes. These latter are as follows: worry about overuse (8 per cent), headache-nausea (4 per cent), other physical distress (for example, sweating, dizziness) (6 per cent), trouble in thinking (4 per cent), insomnia (2 per cent), difficulty in studying (2 per cent), criticism by friends or relatives of use (2 per cent), other psychological distress (inability to be alert when with others) (2 per cent), appetite problems (1 per cent), slurred speech (1 per cent), and unconsciousness (1 per cent). Once again, intensive users report more diverse ill effects than less intensive users; the average for the former is 0.34 per person and for the latter 0.15 per person.

About one fifth (19 per cent) of the students have had experience with marijuana. Seventy-seven per cent report benefits from the drug and 46 per cent report ill effects as follows: headache and/or nausea (21 per cent), sensitivity to criticisms of friends or relatives regarding use (16 per cent), emotional upset due to the drug (15 per cent), other physical distress (11 per cent), trouble in thinking (9 per cent), hallucinations (regarded as unpleasant) (8 per cent), appetite problems (8 per cent), slurred speech (6 per cent), worry about overuse (6 per cent), loss of self-control (4 per cent), incidents leading to trouble with friends (4 per cent), (other) psychological distress, for example, "remoteness" (3 per cent), study problems (2 per cent), jail (2 per cent), bad dreams (2 per cent), insomnia (2 per cent), school discipline arising from drug use (1 per cent), car accidents arising from use (1 per cent), other social problems (1 per cent), unfortunate sexual incidents or difficulties (1 per cent), and criminal acts attributed to the drug-influenced state (1 per cent). Intensive users report more diverse complaints than less intensive users; the former average 1.3 different ill effects per person, the latter 0.8.

Five per cent of the students report experience with hallucinogens. Among them 79 per cent report benefits and 56 per cent report bad outcomes. These latter include unpleasant hallucinations, including recurring ones (23 per cent), emotional upset (22 per cent), headache/nausea (20 per cent), sensitivity to criticisms of friends or relatives about use (17 per cent), insomnia (15 per cent), trouble in thinking (15 per cent), (other) psychological difficulties (9 per cent), slurred speech (8 per cent), other physical disorders (8 per cent), bad dreams (6 per cent), loss of self-control (6 per cent), psychosis (5 per cent), worry about overusing hallucinogens (6 per cent), appetite problems (3 per cent), problems in studying (3 per cent), unconsciousness (3 per cent), and incidents leading to troubles with friends (2 per cent) (one case reporting). In the case of hallucinogens, intensive use is not linked to the report of more diverse ill effects, for the average number of different bad outcomes listed by intensive users is 1.3 compared with 1.6 for less intensive users.

One and three-tenths per cent (N = 18) of the students report illicit opiate use. Among these eighteen students, 56 per cent (N = 10) report benefits and 72 per cent (N = 13) ill effects. Bad outcomes reported include headache/nausea (44 per cent, N = 8), sensitivity to the criticism of friends or relatives (28 per cent, N = 5), worry about overuse (22 per cent, N = 4), troubles in thinking (17 per cent, N = 3), emotional upset (17 per cent, N = 3), other physical disorders (17 per cent, N = 3), slurred speech (11 per cent, N = 2), appetite problems (11 per cent, N = 2), hallucinations (6 per cent, N = 1), problems in studying (6 per cent, N = 1), and car accidents (6 per cent, N = 1). Intensive users report more diverse ill effects than less intensive users; the average for the latter is 1.7 different bad outcomes per person, and 0.8 for the former.

Seventy-nine students (6 per cent) report experience with special substances; 42 per cent report benefits and 42 per cent report bad outcomes. These latter include headache and/or nausea (24 per cent), other physical disturbance (11 per cent), sensitivity to criticism of friends or relatives about use (5 per cent), hallucinations (4 per cent), incidents leading to troubles with friends (3 per cent), appetite problems (3 per cent), trouble in thinking (3 per cent), slurred speech (3 per cent), (other) social problems—for example, finding it hard to be with a group (3 per cent), loss of self-control (r per cent), car accidents (1 per cent), emotional upset (1 per cent), psychosis (1 per cent), (other) psychological disorders—for example, impatience with others (1 per cent), and worry about overuse (1 per cent). There is no difference between intensive and less intensive users in the average number of different bad outcomes; the former average .5 per person and the latter 0.6 per person.

OFFICIAL CASES

We wanted to compare the prevalence of kinds of bad outcomes and the inferred minimal incidence of bad outcomes as reported by students using drugs with official cases—that is, the kinds of troubles and numbers of students in trouble known to school, police, and school medical authorities. The question in our minds was to what extent the cases identified by authorities as being in trouble in consequence of, or in association with, drug use accurately reflected use on campus and actual difficulties as reported by students. During the study year we asked school officials to keep a record of all students in school known to them because of drug-related difficulties. We put the same request to medical personnel on those campuses maintaining a health facility. Kinds and numbers of cans were then reported to us (not by name). At the end of the year, we also asked all state and local law-enforcement agencies concerned with drug offenses whether they had identified any offenders as students from the five schools. The adequacy of all of these records as rendered to us depended on the excellence of the record systems in each office and school. We did not conduct a study of these record systems but presume, generalizing from our work in other agencies (Blum and Ezekiel, 1962), that official record keeping was usually not very good. We must assume that some students who were, in fact, identified officially failed to be recorded as such. The police records failed entirely to report identifications, which was not a fault of their record keeping but rather represented a failure on the part of the police-record systems to inquire about student status.

The schools differed considerably in the kind and detail of records kept on discipline, academic, and health problems. Here we shall attempt to compare, combine, and summarize the differing levels of information as best we can.

In School I, ninety-six disciplinary cases were officially processed. These represented 1.6 per cent of the undergraduate student body. Between six and ten of these cases involved drug use, although whether the drug was alcohol or an illicit substance was not specified in the records. The maximum of ten cases involving drugs and discipline reflected 0.07 per cent of the total undergraduate population. In School II there were fifty-two disciplinary cases representing 1.7 per cent of the undergraduate population. Among these fifty-two cases, thirty (58 per cent of the total) involved rule violations associated with alcohol use-abuse, seven involved both alcohol and other (illicit) drugs, and two involved illicit drugs only. The alcohol-only cases comprised approximately 1 per cent of the student body; the alcohol and illicit drug group were 0.2 per cent and the illicit drugs only 0.05 per cent. For School III we were not given total disciplinary cases but were informed that one student was referred for rule violation in association with illicit-drug use and that five to ten students were disciplined for alcohol-related incidents. These cases (taking the maximum of ten for alcohol) constituted 0.2 per cent and 0.02 per cent respectively. For School IV disciplinary data were not available. For School V we learned that 196 students, 1.4 per cent of the undergraduates, were found guilty of disciplinary violations. Among these, alcohol-related offenses constituted seventy-one cases (36 percent of all those disciplined) or 0.5 per cent of the total student body. There were no cases involving illicit drugs.

Summarizing the discipline data, we find that official discipline cases apparently involve less than 2 per cent of the students in any school, that drug cases of any sort comprise from 5 per cent to 75 per cent of all disciplinary cases, and that alcohol-related violations exceed violations involving any other drugs by a factor of at least ten—that is, there are at least ten students identified as rule violators in cases involving alcohol use for every case involving another drug.

Arrest data were available for two schools only, and for these they were inadequate. In School I, two students, 0.03 per cent of the full-time student population, were known to have been arrested. In School II no arrests were known; for School III and School IV there were no data; and for School V there were five arrests, comprising 0.04 per cent of the undergraduate student body. Summarizing these data, we can report that not many student arrests are evidently known to school authorities or to reporting police agencies, although all concerned agree that student arrests are likely to have occurred which have not been identified as such.

In School I, 71 students were suspended or disqualified ,for academic reasons and another 240 were put on probation. Of these 311 cases constituting 5.3 per cent of the undergraduate population, 30 to 45 were identified by deans as students believed to have drug-use (not alcohol) problems. No evidence was available to deans as to the causal role of drugs in these cases which constituted (taking the maximum, forty-five) 19 per cent of all academic-performance problem cases and 0.7 per cent of the undergraduate student body. In School II, we were not given total academic-failure data but were informed that five academic cases were also identified as having alcohol problems. Each of these cases also appeared as a discipline violation. These represented 0.02 per cent of the students. Schools III and IV provided no data on academic problems. School V provided na data on total academic-performance problems, but did report that fifty-one students identified as users of illicit drugs by deans failed, were put on probation, or dropped out. These represented 0.4 per cent of the undergraduate student body. We conclude that a maximum of 1 per cent of the students, and usually far fewer, will be identified as drug-use problems associated with difficulties in academic performance.

School I has an active student-health service. During the study year it reported no adverse drug reactions during the acute period as cases. It did report that seven students with adverse drug reactions (LSD, DMT, and so on), coupled with broader personality disorder, did present themselves as cases as their acute reactions had passed. The health service also saw one student for the treatment of alcohol-precipitated distress. These cases represented about 1 per cent of all psychiatric visits to the health service and represented a rate of prevalence of drug-related disorders (including alcohol) among health-service cases as a proportion of the total student body of about 0.07 per cent (.00007).

School II had no adverse reactions to report. School III does not operate a health service, but at the beginning of the study year, it did provide a psychologist-counselor to see students with drug problems. There were fifteen to twenty self-referrals to that counselor during the first months of the school year, a group comprising 0.4 per cent of the student body. That counseling service was closed when the counselor was assigned other duties; there was also evidence at the time that students were becoming reluctant to make visits, some saying it was for fear of being reported to the police under the thenjust-passed California felony law for LSD.

School IV reported 30,000 health visits or about 2.4 visits per student. Of these, 306 individual cases were seen by psychiatrists (2.5 per cent of the students) of whom 11 were LSD reactions (3.6 per cent of all psychiatric cases seen, 0.09 per cent of all students). School V showed 628 students seen by psychiatrists (4.5 per cent of all undergraduates), of whom 6 had adverse illicit-drug reactions and 8 had drug use as a feature of personality difficulty. The adverse reactions constitute 0.04 per cent of all students and about 1 per cent of the psychiatric cases seen. On the basis of these data, we conclude that adverse drug reactions as such constitute less than 4 per cent of all psychiatric cases seen and that fewer than one tenth of one per cent of the student body are identified by health services, on any campus, as having adverse reactions.

DATA AND STUDENT REPORTS

The official data allow prevalence counts for the year 1966— 1967, whereas the student reports are by drug and cover lifetime prevalence and do not allow specification of year. Thus, any comparison is bedeviled by these different bases and can be, at best, only a very rough order comparison.

The simplest statement, given the considerable amount of use of the approved drugs—alcohol, sedatives, and so on—and the prevalence of physical ill effects described, is that few or none of the ill effects from sanctioned drugs come to the attention of health authorities.1 Thus, while half of the total sample have experienced physical distress attendant upon alcohol use—and their drinking is continuing and common—no case of intoxication, hang-over, and so on is officially known. Only alcohol, of the approved drugs, ever leads to bad outcomes of any kind identified by school authorities.

To assess the relationship of official statistics to bad outcomes, it may be helpful to report these latter not as a percentage of students using a particular drug but as a proportion of the total population, since the official statistics use as their sample that total population. Consider that 58 per cent of all students in our sample report bad outcomes from drinking—that is, over half of all students on the five campuses have defined in themselves some ill effects following drinking. One presumes that many students have had repeated bad outcomes of diverse kinds. Consider further that 138/1,314 are worried about their own excessive use (which we take to be dependency fears —that is, awareness of compulsive drinking), that 143 have suffered loss of self-control, 65 have been in fights, 38 have alcohol-related study problems, 35 have had auto accidents, 35 acknowledge they have received school discipline, and 20 students (20/1,314) admit having gone to jail for alcohol-related offenses. Yet the highest school figure—that from School II with its excellent individual discipline records specifying drug problems—shows only 1 per cent of the total student body as having had drug troubles (alcohol included) during the study year. Some of the difficulties reported by students may have occurred in prior years but, as we shall see, bad outcomes for alcohol become more prevalent with increased age and school year (as drinking increases). This means that the chances for a bad outcome's having occurred during the student's tenure in the school—and during recent years especially—are quite great. As a general conclusion, then, we must state that even the most carefully kept school records represent only a fraction of the unpleasantness and distress which students experience in association with their use of the approved drug. Put another way, it can be assumed that students handle that distress either privately—gritting their teeth—or informally, getting assistance from friends or possibly—and we suspect rarely—family physicians. We have no way of knowing to what extent parents or family physicians (and keep in mind many students are away from home so they cannot consult the family doctor, and so their medical care should appear in health-service records) are aware of student difficulties in association with sanctioned drug use; for example, are the parents aware of the suicide attempts which two students said occurred during a drinking episode? We shall, in a study now planned, attempt to establish the amount of information held by parents and physicians about their children's and patients' level of drug use and drug difficulties.

The same problems which we infer to exist in relationship to alcohol use occur with the other approved drugs.2 There the schools report no data, and student prevalence reports of difficulty are less than for alcohol, but, nevertheless, troubles do exist. For example, 268/1,314 students are worried about their tobacco habit and would presumably be open to assistance aimed at reducing their smoking, certainly a worthwhile endeavor, given our knowledge of cancer and heart disease associated with smoking. Similarly, twenty-two students are concerned about their own possible amphetamine dependency and twenty-five about sedative dependency. Who knows of their fear and who can help?

What about illicit drugs? Even though the approved drugs are used by more students, and alcohol has the highest prevalence of different ill effects reported for intensive users, the illicit drugs—with their less frequent employment (except in samples of extreme illicit-use groups, see Chapter Eight )—are not without bad outcomes. Consider, for example, that the hallucinogens among all of the drugs have the highest prevalence among less intensive users of bad outcomes, half again as great as alcohol. Consider, too, that one particular bad outcome is potential for each incidence of use of each of the illicit drugs—and that is arrest. Given our finding that almost none of the physical and psychological disturbances in association with use of the approved drugs become official cases, we would expect little official reporting of disapproved cases because of the problem of detection, shame, and so on. On the other hand, anxiety about the use of these drugs must be predicated so that individual students suffering ill effects would be expected to be more anxious about them and thus conceivably propelled into treatment.

Again, to facilitate comparison of official total-population figures with bad outcome reporting, we present bad results for illicit drugs as a per cent of the total population of students rather than as a per cent of using students. According to that scheme, a number equivalent to 4 per cent of all students suffer headache/nausea attributable to marijuana, 1.4 per cent have undesirable hallucinations, and 2.8 per cent emotional disturbance attributed to marijuana. About 1 per cent are concerned about their own marijuana dependency. Among hallucinogen effects, about 1 per cent suffer emotional upset or undesirable hallucinations, four students are worried about dependency, and three (0.2 per cent) say they have suffered psychotic reactions (to be distinguished from psychotomimetic effects or the acute toxic period). It is the academic performance rather than discipline records which best identify student illicit-drug users; for example, never more than 0.07 per cent of the students at any school were identified, during the study year, as illicit-drug users involved in a discipline problem—academic reports showed ten times greater an incidence rate, up to 0.7 per cent—which is still less than 1 per cent. Health or counseling records, on the other hand, also never achieved identification of more than one tenth of one per cent as cases of adverse effects among illicit users.

If we consider that the average age for initiation of illicit use is between nineteen and twenty-one for all users, we must assume that most of the bad effects reported have occurred during the college years and that some of them will have occurred in the study year—especially since the average age for the total sample is twenty years. If we assume, conservatively, that 1 per cent of the sample did suffer at least one bad reaction during the study year and we compare that with official cases, we see that health and discipline records bear a relationship—if that is what it is—of never more than one identified case for every fourteen self-identified cases. On the other hand, academic performance estimates can be closer, for every case of a self-identified bad outcome—very few of which are ever reported in fact as study problems—the maximum official rate was 0.7. These figures are only an exercise based on assumptions easily—and quite wisely—challenged. With data of the sort available we can do no better.

PREVALENCE OF USE

We have concentrated on students' reports of bad outcomes in comparison with official statistics relating drug use to problem cases. The larger question is what relationship does knowledge based on official cases have to the prevalence of illicit-drug use on campus? One can also ask how accurate are the estimates of illicit use which are arrived at by means other than official case finding, as, for example, the estimates offered by students themselves?

We think it requires little further discussion to affirm that at the time this study was made and on the campuses where it was made, illicit-drug experience was widespread, although still confined to a minority of students. Chapter Eleven shows that during the year of the study—and in the following years while our data were being analyzed—there was an increase in use—that increase amounting, at our best guess, to a two- or three-fold jump in the number having marijuana experience, that is, up to an average 40 per cent to 60 per cent. What relationship officially identified cases of trouble—discipline, academic, arrest, or health—had to prevalence of experience is very clear indeed. Official identification, including all official measures, never achieved a rate of identification of over 1 per cent of the students as illicit-drug users. Official case identification was often far below 1 per cent. We conclude that official cases do not provide any estimate of actual illicit use on campus and that official cases as processed on these campuses—including arrest figures by police—ordinarily are underreported by a factor of from 1/20 to 1/1,000.

Official cases, of course, represent more extreme reports either of persons or of institutions. Most things are handled informally by institutions; most conduct is not officially noticed. This does not mean that people—whether deans, doctors, detectives, or students—are unaware of what is going on around them; they, too, are part of an ordinary unofficial world which provides daily evidence—albeit based on limited samples—of what is happening. So it is that people do make estimates of the prevalence of things; and one of the things they estimate is the prevalence of illicit-drug use. As we said in the introductory chapter, when comparing estimates made by the journalist Richard Goldstein (1966), those seat-of-the-pants figures can be rather good approximations. Because we are interested in the general accuracy of such estimates and particularly in the accuracy of illicit-drug users compared with non-users, we now turn—at a tangent for the moment—to a consideration of the accuracy of drug-use estimates by students themselves.

We expected that the more students themselves used illicit drugs, the higher would be their estimates of others' illicit use. The expectation was based on their selectivity in acquaintances, the fact that (Blum and Associates, 1964) drug users associate with each other more than with non-users and, also, that they can more easily identify—through their gossip channel—other users. We also expected they would want to believe there were many users because there is discomfort if one feels alone in being unlawful and deviant, a sentiment we think is linked to that proselytism which so characterizes illicit-drug users (Blum and Associates, 1964) . For the sake of completeness, we shall not only compare estimates by intensive and less intensive users of illicit substances, compared with each other and with non-users, but we shall also note estimates by class of users of the other drugs.

First, applying the Kolmagorov-Smirnov Test, we learn that there are significant differences between the estimates of nonsmokers compared with lesser smokers, as well as between lesser smokers versus intensive smokers. The same significant differences both between non-users and users and between less intensive users compared with more intensive users obtain for alcohol and marijuana. Significant differences between non-users and either intensive users or less intensive users, or both, are found to hold for amphetamines, sedatives, tranquilizers, hallucinogens, and special substances.

Actual illicit experience varies by school. We can, at best, only estimate it since there is overlap by drug class and since the illicit use of amphetamines, except by first source, by 55 per cent does not allow us to estimate later illicit amphetamine use. It is also not possible for us to distinguish between purchase on the black market, a frankly illicit act, and informal uses which would not be prosecuted even if apprehended—as, for example, taking an amphetamine tablet from a roommate. Therefore, if we limit our base figure of illicit use simply to the per cent of students with reported marijuana use (since all but five hallucinogen users say they have used marijuana) plus a 5 per cent to 10 per cent arbitrary "safety" factor, we can then determine which groups classified by use of each drug make the best estimates on each campus of any illicit experience. A measure which is likely to be more sensitive to the mood of the students is one which uses as a base not estimates of persons with past illicit experience but one which estimates student intentions either to continue marijuana use or to begin it.

School I has actual marijuana experience of 21 per cent but shows that 32 per cent intend to begin or continue, with another 9 per cent unsure. In contrast, the student-body average estimate of illicit experience is in the range of 10 per cent to 19 per cent, which is low. Those who make estimates in what we take to be the correct range of actual experience, 20 per cent to 30 per cent and defined as the median estimate group, are low-intensity tranquilizer users, low-intensity amphetamine users, high-intensity marijuana users, high-and low-intensity hallucinogen users, high- and low-intensity opiate users, and high-intensity special-substance users. The groups sensitive to the intentions of the students (and this is the campus for which we have some evidence of considerable drug increase in the two years following the initial survey) are high-intensity amphetamine users and the few low-intensity opiate and special-substance users. It seems clear that on this campus, a large private university, the majority underestimates illicit-drug experience and that almost all groups underestimate the potentials for increased drug experience. Those who come closest to being accurate are illicit-drug users but even they underestimate the reserve of students ready to "turn on."

On Campus II, a Catholic university, marijuana experience is reported by 11 per cent. If we add our arbitrary safety factor for illicit use, it would remain in the range of 10 per cent to 20 per cent. Positive intentions to begin or continue use are expressed by 24 per cent, with another 1 per cent not sure. The average estimate by the total student body for illicit use on campus is between zero to less than 9 per cent in range, which is obviously low. Those groups correct in their experience estimates are the intense tranquilizer users, along with the less intensive marijuana users. No groups successfully estimate the mood of the campus when defined as the greater number who intend to continue using illicit drugs or to try them.

Campus III, the junior college, reports 21 per cent marijuana experience which, with the safety factor added, remains in the 20 per cent to 30 per cent range. Twenty-one per cent also intend to continue or begin, with an additional 11 per cent uncertain. For this campus we deem intention to remain within the 20 per cent to 30 per cent range. The average estimate by the total student body for the per cent of students experienced in illicit drugs is in the 10 per cent to 19 per cent range, although the modal estimate is in the 20 per cent to 30 per cent range. The student groups who make accurate estimates of experience and intentions are tobacco smokers, heavy drinkers, low-intensity amphetamine users, low-intensity sedative users, (all) tranquilizer users, and the few intensive special-substance users. On this campus we encounter overestimates, all of which are by marijuana and hallucinogen users.

Campus IV suffered, as noted in the introductory chapter, from poor sampling, so that our estimates of drug use here are' very likely to be in error. Since sampling bias arose out of overpresentation of older upperclassmen, a group already shown to have greater drug use than others, the reported experience here must be assumed to be erroneously high, that is greater than the actual total-population experience. On this campus, marijuana experience is 33 per cent; our safety factor would raise that but keep it in the 30 per cent to 49 per cent range (the same range as used in coding estimates). Intentions to begin or continue marijuana use are expressed by 34 per cent, with another 9 per cent uncertain. We deem the range unchanged-30 per cent to 49 per cent. Actual estimates by the total student sample yield a mean of 20 per cent to 29 per cent, although the modal estimate is in the 30 per cent to 49 per cent range. If we are correct in describing the sample as biased toward illicit-drug experience, then the overall student average estimate would be correct. Student groups achieving estimates in the 20 per cent to 30 per cent range include nonsmokers and low-intensity tobacco smokers, light drinkers, non-users of amphetamines, sedatives, tranquilizers, marijuana, and the other illicit drugs, as well as less intensive sedative users. Those groups arriving at the higher figures, which are the ones reported by our sample, are heavy tobacco smokers, heavy drinkers, ( all) amphetamine users, (all) hallucinogen users, (all) tranquilizer users, less intensive marijuana users, heavy sedative users, and less intensive special-substance users. Overestimates are given by intensive marijuana users, (all) opiate users, and heavy special-substance users.

Campus V, a large state university, reports actual marijuana experience by 10 per cent of the students. Intentions to begin or continue use are expressed by only 5 per cent of these users. In this school a coding error occurred with the result that no non-users were asked about their intentions, so that we cannot gauge their readiness to begin. If we assume that at least 6 per cent to 15 per cent are in that group, then the range for future users is 10 per cent to 20 per cent. The average sample estimate of experience is in the less than 0 to 9 per cent range. Groups correctly estimate actual experience and our assumed future-users group, both in the 10 per cent to 20 per cent range, when they are intensive marijuana users and less intensive hallucinogen users. No overestimating groups were found. On this campus there is a consistent trend to underestimate illicit-drug use slightly.

Estimates of how many students on campus use drugs illicitly are significantly associated with students' own use of all classes of drugs; the more the drug use, the generally higher the estimate. The accuracy of estimates varies from one campus to another and may be measured either in terms of figures derived from past experience or from statements of plans for use, which constitute, we believe, a greater sensitivity to the campus mood or readiness to take drugs illicitly. On the majority of our campuses, non-users of drugs and the student body as a whole are consistently low in their estimates; on two campuses only are overestimates of use in evidence; these were made by marijuana users primarily. No one student group, classified by its drug use, is accurate for all campuses. What stands out are (1) the general underestimates of past use and especially of readiness for future use and (2) the need not to discount the estimates by the illicit users which, though sounding high, may very well be correct.

CHARACTERISTICS OF BAD OUTCOMES

We now return from our tangential discussion of the estimation of illicit-drug use back to the primary concern of this chapter: bad effects. At this point we can ask whether there are any outstanding characteristics among drug users who report bad outcomes as compared with users who do not make such a report. As in earlier chapters, we shall report trends. At the outset, remember that we have already shown that for every drug class except the hallucinogens—and to a lesser extent the special substances—a greater number of diverse bad outcomes are reported by those who use drugs intensively than by less intensive users. Remember also from Chapter Six that intensive users are, for the most part, more extreme in their characteristics than less intensive users and are likely to be (although not always so) a mid-point on a theoretical continuum. We have already shown in Chapter Three that non-users of any drug are a group constituting the anchor of the continuum at the other extreme.

The characteristics of those reporting ill effects compared with those not reporting them are not nearly as consistent, either by characteristics in relation to general drug-use trends as we saw in Chapters Three and Four or in relation to intensive users vs. less intensive ones as reviewed in Chapters Five and Six. Part of the problem is the small N in some cells as one gets down to particulars, such as "those who are marijuana users who have used drugs for suicide efforts and who do report adverse drug effects." We suspect a larger problem is that, other than its demonstrated relationship to drug intensity, except for hallucinogens and special substances, the bad-effect variable is compounded of many features whose extent and ramification's are quite beyond the scope of the present endeavor. Bad effects, to be identified and reported, cover the gamut of social, psychological, and physiological processes. In spite of its inconsistencies and surprises, we present this limited material, which should be considered primarily as a basis for speculation and future research.

We limit ourselves to reactions to four drugs of interest: alcohol, amphetamines, marijuana, and hallucinogens. Our first observation is that the ill effects of alcohol are mostly reported by people with the same characteristics that predict drinking in general and heavier drinking in particular. One surprising feature is that the politically left-of-center students are more heavily represented among those having ill effects than would be expected from simple trend data on use or use intensity. Another surprising feature is that Jewish students—in spite of statistical overrepresentation on the left—have the lowest rate for any religious group of reported ill effects.

Notable among characteristics associated with amphetamine adverse effects or incidents is the low prevalence of bad outcomes among students from wealthier families, from Jewish families, among those politically active, among Republicans, among those recalling little parental concern over their health, and those dissatisfied with school and course work. If these characteristics serve to identify a special subgroup protected from ill effects, then others suggest those particularly vulnerable. We find that females, older students, those from unsettled families, those politically left of center and those with a history of dropping-out and planning to drop out remain proportionately more at risk of ill effects; we say "remain" since we recognize these characteristics as ones linked to drug experience and intensity.

Marijuana ill effects are notable primarily for the apparent reversals in characteristics ordinarily linked to use and intensity of use. At higher risk of adverse outcomes are females, younger students, lowerclassmen, those from poorer families, those in agreement with their fathers' politics, those whose parents did worry about their childhood health, and those who do find their studies related to how they want to live in the future. On the other hand, what may be another more familiar group "at risk" are those with childhood-illness advantages, those with incomplete grades, those who are school drop-outs, those who are pessimistic about doing what they want, and those using drugs for diverse functions.

With reference to hallucinogens, our data provide no insights into those experiencing adverse outcomes. One finds few notable reversals or dramatic trends. On an impressionistic basis, supported both by the evidence for no increase in ill effects with increases in hallucinogen use and by the hodgepodge of characteristics linked to it, we would propose that adverse hallucinogen effects—and these are, for the most part, LSD effects in the student sample—are less linked to pgychological, social, or ideological variables identified in this study than are adverse effects from other drug classes. If such is the case, it would follow that other or "deeper" psychological variables are involved—as, for example, a prepsychotic personality—that the life situations of these LSD sufferers are so bizarre as to be ruled social variables outside of the ordinary ken, or that the outcomes noted are determined in part by biochemical idiosyncrasies not linked to the kinds of characteristics with which this study concerns itself. LSD bad outcomes are, we propose, a different kettle of fish.

SUMMARY

Examining the prevalence of kinds of bad outcomes reported by students from their drug use, as well as benefits claimed, we find that ill effects are described for all drugs used. The prevalence of ill effects as a proportion of those reporting them over the total population using the drugs is greatest for the opiates and for alcohol, followed by amphetamines and hallucinogens, and least for sedatives and tranquilizers. These rates do not imply incidence—that is, they do not say how many times a person has had ill effects or what the expected rate of ill effects per occasion of use may be. Since we assume that alcohol-use occasions are likely to be more frequent than any other while hallucinogen use is likely to be quite infrequent, we can surmise that the risk of ill effects per occasion of use is much greater for hallucinogens. As for benefits, these are claimed by the majority of users for most drugs; fewest benefits are claimed for opiates and the special substances. Kinds of ill effects vary by drug; physical distress, for example, is most often mentioned in regard to alcohol, as are accidents, fights, criminal offenses, and so forth. With amphetamines, emotional upset is the most common complaint; with marijuana and opiates, headache/nausea and discomfort over criticism from others rank foremost, and with hallucinogens, unpleasant hallucinations and emotional distress are foremost. Psychoses following use are noted with the hallucinogens more than any other drug.

Examination of official cases reveals that school records, whether academic, disciplinary, or medical, report far fewer cases of distress than students themselves report. Indeed, almost none of the ill effects from approved social or prescribed drugs appears to come to the attention of school authorities. Police records, which are not geared to identifying students as having ill effects, are not a source of student-arrest data. Discipline and health records never achieve a ratio of identification of bad outcomes greater than 1 to 14 as compared with population self-reports. However, careful examination of academic performance, including drop-outs and probations involving student illicit-drug use, can achieve an identification-rate estimate of students reporting drug problems which is only a little less than the actual rate of occurrence. That the number of students identified as having academic problems associated with illicit-drug use can approach self-reported prevalence rates for bad outcomes in no way gives an estimate of the incidence of bad outcomes or of the simple prevalence of students on campus with illicit-use experience. On the other hand, attention to estimates of use per se rather than bad outcomes leads us to conclude that most estimates made by most students are low; if one asks the estimates of use to encompass those intending to use drugs—many of whom, we believe, will become or have become users—the estimates are even less adequate. The exceptions tend to vary by campus but users of illicit drugs more often give better estimates than do non-users. On any campus overestimates of either use or of intentions to use are rare, even when those estimates are made by intensive users of illicit-exotic drugs.

Analysis of the characteristics of students reporting bad outcomes shows the expected link between the amount of use and the prevalence of bad outcomes. This leads to the inference that characteristics associated with intensive use are also the characteristics associated with the bad outcomes reported. The most important exception is in the case of hallucinogen use; here, infrequent one-time users report more ill effects than frequent users. Other exceptions include Jewish students, who are low in bad outcomes for alcohol and amphetamines. In one subgroup marijuana ill effects are apparently linked to special features not part of the ordinary intensive-user trend; specifically, females, younger students, lowerclassmen, poorer students, and those closer to family report more distress. Intensive users with the usual constellation of dissatisfaction, pessimism, and so on suffer equivalent ill effects. These characteristics which appear linked to outcomes are, for the most part, not striking. In the case of hallucinogens—where there is that unusual phenomenon of low use but high ill-effect prevalence—one senses a special case that can be accounted for by positing biochemical idiosyncrasy, individual psychopathOlogy "setting off" acute reactions triggered by single drug use, progressive loss of critical faculties with increased use, or altered definitions of what constitutes an ill effect as use increases. Sheer sampling artifact and unwarranted inference on our part may also be occurring.

1 We could not survey private physicians and hospitals.

2 Keep in mind that considerable amphetamine use is illicit but is classified here under the approved drugs.