The Epidemiology of Multiple Drug Use


How much multiple drug use is there? What proportion of the population at any one point in time is using/abusing multiple substances? Has use of multiple substances become more normative in the recent past as opposed to exclusive use of a favorite drug? What are the principal consequences of multiple drug use? Do these consequences differ according to pharmacological parameters for interactive potential or are there other parameters of almost equal predictive value? To what extent are the consequences attributed to single drugs (traffic accidents labeled as alcohol related) really the result of impaired judgment and performance from ingestion of multiple substances? These are just a few of the questions that need to be addressed within the scientific and public policy communities. In the following section some epidemiological data pertinent to understanding the “extent” of multiple drug use are presented.

Monitoring-the-Future Studies

Each year since 1975, researchers at the Institute for Social Research at the University of Michigan have administered questionnaires to about 17,000 high school seniors attending schools randomly chosen to be representative of all high schools in the continental United States. These seniors are asked to provide detailed information on their use of various drugs in addition to attitudes/opinions/beliefs/observations on a host of topics. There are five separate forms of the questionnaire with some core content shared by all. The study does not include persons who dropped out of school prior to their senior year or persons who are absent on the day the questionnaries are administered. It is estimated that the dropout rate in the United States may average 20-25% of a birth cohort. It is also known that drug use among dropouts is higher and more extensive than among their peers who continue in school. However, in an analysis of how noncoverage of these populations would affect the prevalence and incidence rates for drug use, Clayton and Voss concluded that the effect would be relatively small. A high rate of use in a small proportion of the population has only minimal effects on overall rates for most drugs. Data from the Monitoring-the-Future studies are therefore conservative estimates of the “true” scope of the problem. Stated differently, because the same measures have been used on similar populations every year, the configuration of the slope for drug use is very reliable, and only the height of the slope for each year is off because of the populations not covered.

Has multiple drug use among high school seniors increased? One way to answer this question is to examine changes in the percentage of youth who have used “only marijuana” during the past year compared to those who have used other illicit drugs. As noted earlier, use of marijuana seems to be a prerequisite for use of other illicit drugs. Furthermore, Akers notes that “… while a great many young people drink without using marijuana, drinkers are more likely than non-drinkers to smoke marijuana, and almost all marijuana users drink.” Therefore, even the “marijuana-only” users (defined solely in terms of use of illicit drugs) are usually multiple drug users because of their use of alcohol. With this caveat in mind, the data in Table Multiple Drug Use in Past 12 Months: Ratio of Use of Marijuana Only to Use of Other Illicit Drugs by High School Seniors show the “annual” prevalence rates of any use of an illicit drug, use of only marijuana, and use of other illicit drugs (the implication is that these youth have used or also use marijuana) for each high school senior class from 1976 through 1983. Simple visual inspection reveals that any use of illicit drugs increased from 48% in 1976 to 54% in 1978 and 1979 and since then has dropped to 49%. The range of increase-decrease is not large. The figures for marijuana only went from 23% in 1976 to a high of 27% in 1978. Since then, the percentage of high school seniors who used only marijuana during the preceding year has dropped steadily until it was 16% in 1983. In contrast, the column for other illicit drugs (e.g., any use of hallucinogens, cocaine, heroin, or any use not under a physician’s order of other opiates, stimulants, sedatives, or tranquilizers during the preceding 12 months) reveals an increase from 25% in 1976 to 34% in 1981 and 1982. In 1983, this figure was 33%. Simply put, at least 33% of the class of 1983 could be classified as multiple illicit drug users.

Table Multiple Drug Use in Past 12 Months: Ratio of Use of Marijuana Only to Use of Other Illicit Drugs by High School Seniors

Classof Any use of illicit drugs (%) Marijuana only (%) Other illicit drugs (%) Ratio: marijuana only divided into other illicit drugs Ratio by sex and by college plans
Sex College plans
Male Female No Yes
1976 48 23 25 1.08 .96 1.26 1.32 .91
1977 51 25 26 1.04 .90 1.14 1.25 .81
1978 54 27 27 1.00 .90 1.13 1.15 .82
1979 54 26 28 1.08 1.00 1.08 1.28 .89
1980 53 23 50 1.30 1.15 1.50 1.71 1.08
1981 52 18 34 1.89 1.57 2.00 2.11 1.58
1982 51 17 34 2.00 1.60 2.50 2.24 1.61
1983 49 16 33 2.06 1.78 2.20 2.31 1.65

Perhaps a better way to illustrate the changes in multiple drug use with this sample would be to create a score for multiple illicit drug use compared to use of only marijuana. This can be accomplished by dividing the marijuana-only percentage into the percentage reporting use of other illicit drugs, the implication being that they were also using marijuana. The resulting ratio score reflects multiple illicit drug use during that year.

The ratio of “marijuana-only” users to those who had used “other illicit drugs” during the preceding 12 months was 1.0 in 1978. As one reads down the columns of ratio values in Table Multiple Drug Use in Past 12 Months: Ratio of Use of Marijuana Only to Use of Other Illicit Drugs by High School Seniors, it is clear that multiple illicit drug use has increased. For the class of 1983, the percent who reported use of “other illicit drugs” (and by implication, marijuana as well) was more than twice as large as the percent reporting “marijuana only” during the preceding year. This pattern of increase in the proportion of high school seniors who are multiple illicit drug users appears also to have occurred for both men and women and among those who plan to go on to college and those who will not go on.

Although somewhat crude, and limited to just the past 12 months, this ratio measure suggests that there has been an increase in multiple illicit drug use among high school seniors since 1978 or 1979. The slope of the increase curve seems to be rather steep and seems to have basically the same configuration for both sexes and for college bound as well as non-college-bound seniors. However, the ratio assesses multiple drug use in macrolevel terms and does not seems to address the questions of simultaneous use of multiple substances or the number of persons at risk for problems because of combined use of multiple drugs.

The Monitoring-the-Future surveys contain essentially the same questions each year. Some of these questions require the respondents who have used a drug during the preceding year to estimate the number of times their use of that drug coincided with use of other drugs so that the effects would have overlapped. The exact question is: How many of the times when you used (drug) during the last year did you use it along with (drug) — that is, so that their effects overlapped? The response alternatives are: not at all, a few of the times, some of the times, most of the times, every time. Clearly, those who claim simultaneous use “most of the time” or “every time” are multiple drug users in the most rigorous definition of the term. From a prevalence/incidence epidemiological perspective, it should be remembered that the subset of persons reporting combined use of a pair of drugs on most or all occasions during the past year are really a part of a part of the overall sample. This will be illustrated in the description of the data found in Table III taken from the classes of 1976, 1978, 1980, and 1982.

Reading across the first row of Table Table Multiple Drug Use in Past 12 Months: Ratio of Use of Marijuana Only to Use of Other Illicit Drugs by High School Seniors, 41% of the Class of 1978 reported having used marijuana or hashish during the preceding 12 months. Another way of saying this is that 41 out of 100 high school seniors in 1978 reported having used marijuana during the prior year. Twenty-one percent of that part of the class of 1978 reported using alcohol most of the times or every time they used marijuana. Therefore, 8.61 out of 100 seniors in the class of 1978 (41 x .21) used marijuana and alcohol in combination almost every time they used marijuana. A quick perusal of the other figures for the marijuana-alcohol combination reveals very little difference across the four classes — about 9 out of 100 seniors in each class reported use of alcohol on virtually all of the occasions when marijuana was used.

The interpretation given these data depends upon one’s orientation and interests. Someone concerned about changes that reflect an increase might say that the number of high school seniors reporting simultaneous combined use of alcohol and marijuana most or every time they used marijuana has not changed much since 1976. Someone concerned about highway traffic safety and the magnitude of the problem of automobile accidents among youth would be very alarmed. All or almost all of the youth in this study are old enough to have a driver’s license. Close to 9 out of every 100 said they used alcohol and marijuana in combination most of the times or every time they used marijuana during the past 12 months. These are therefore conservative estimates of the number of high school seniors “at risk” to have accidents resulting from impaired judgment and motor performance because of multiple drug use. It is certainly clear from these data that multiple drug use (alcohol with marijuana) may be normative for a sizable minority of the high school senior population.

The data for combined use of marijuana with cocaine and alcohol with cocaine are presented to illustrate the problem of continued use of drugs begun earlier in the developmental process, often in combination with drugs that appear at later developmental stages of drug involvement. Among those who used cocaine in the past 12 months, use of marijuana with cocaine is normative for a relatively large percentage. It is likely that those falling in the combined cocaine-marijuana or cocaine-alcohol categories of multiple drug use are a subset of those who reported combined use of alcohol with marijuana. In analyzing data from the class of 1980, Clayton and Ritter found that 86% of those classified as “daily” marijuana users (used on 20 or more occasions in the past 30 days) had used illicit drugs other than marijuana (e.g., cocaine, amphetamines, tranquilizers) during the past 12 months, in addition to use of cigarettes and alcohol. Daily marijuana users are usually multiple drug users. However, there is a tendency to direct attention to specific, pharmacologically defined classes of drugs with which youth are over-involved. This tendency sometimes leads us to infer that the correlates, causes, and consequences of drug use are attributable to the class with which persons are overinvolved, such as marijuana. If daily marijuana users are more likely than those who use marijuana less frequently to have automobile accidents, be absent from school or work, or exhibit characteristics of the amotivational syndrome, we are prone to attribute these consequences to their use of marijuana. As Clayton and Ritter note: “A more plausible inference is that these behavioral characteristics may be emergent from the ingestion of multiple substances which may be interacting pharmacologically to produce the effects/consequences. ”

National Survey on Drug Abuse

The National Survey on Drug Abuse has been conducted seven times, the most recent being 1982. It is a study of drug use in a representative sample of the household population. In 1982 there were 5624 respondents distributed as indicated in Table “Extent of Use of Other Illicit Drugs within Categories of Extent of Use of Marijuana by Age Group”. The data shown in Table “Extent of Use of Other Illicit Drugs within Categories of Extent of Use of Marijuana by Age Group” reflect the increasing probability of using multiple substances as one moves from less to more extensive use of marijuana, regardless of one’s age group. These data do not refer to simultaneous combined use of multiple drugs. However, the 1982 National Survey interview does contain questions about the frequency (e.g., usually, about half the time, occasionally, rarely, never) with which various pairs of drugs have been used with marijuana.

Table Frequency with Which Marijuana and Cocaine Are Used in Combination for Those Who Have Ever Used Both Drugs by Lifetime Extent of Use of Both Drugs

Lifetime extent of use (times) Usually or about half the times Occasionally or rarely  Never  Total (N)
Marijuana Cocaine N % N % N %
1-2 1-2 2 20 0 8 80 10
1-2 3-10 1 33 2 67 0 3
1-2 11-99 0 1 50 1 50 2
1-2 100 + 0 0 0 0
3-10 1-2 4 10 5 13 30 77 39
3-10 3-10 3 14 7 32 12 55 22
3-10 11-99 0 2 29 5 71 7
3-10 100 + 0 0 2 100 2
11-99 1-2 29 35 14 17 41 49 84
11-99 3-10 14 21 27 40 27 40 68
11-99 11-99 9 24 18 47 11 29 38
11-99 100 + 0 1 100 0 1
100 + 1-2 39 41 16 17 41 43 96
100 + 3-10 95 57 52 31 19 11 166
100 + 11-99 99 64 48 31 8 5 155
100 + 100 + 37 71 9 17 6 12 52
Totals 332 45 202 27 211 28 745

The data in Table Frequency with Which Marijuana and Cocaine Are Used in Combination for Those Who Have Ever Used Both Drugs by Lifetime Extent of Use of Both Drugs show the lifetime patterns of extent of use of marijuana and cocaine (1-2 times, 3-10 times, 11-99 times, 100 times or more) for those who have ever used both drugs along with the reported frequency with which the two drugs have been used together, presumably to either enhance the effects or as part of a normative pattern of use. Several things are clear from these data. First, multiple use of these two drugs is clustered primarily among those who have used marijuana 100 or more times, the highest category of use identified in the interview instrument. Second, among those who have used marijuana 100 times or more, the percentage reporting combined use with cocaine usually or about half the time goes up linearly with extent of use of cocaine. Seventy-one percent of those who have used marijuana and cocaine 100 times or more report combined use is a frequent occurrence. Third, only 28% of the persons in this nationally representative sample of households who had ever used marijuana and cocaine report having “never” used these drugs in combination. Finally, it is important to put these figures into epidemiological perspective. The 745 people who have ever used both drugs constitute 13.25% of the total sample of 5624. The 45% who report combined use usually or about half the time represent 5.96 persons per 100 in the total population (13.25 persons x .45 = 5.96). It should be noted that this is a rather conservative estimate because the denominator initially contained the entire sample of 5624, including youth as young as 12 and 13 and older adults in their fifties, sixties, and seventies. The data from this survey indicate that the preponderance of use of marijuana and cocaine occurs in the 18-34 age range. The number of persons per 100 in this shortened age range who usually or about half the time use marijuana and cocaine in combination is bound to be considerably higher.

It should be noted again that the data from the Monitoring-the-Future surveys and the National Survey on Drug Abuse come from the “normal” population. The respondents in these studies have not been filtered through the institutional or community service agencies where persons with alcohol and drug problems would be clustered. For this reason, the data presented above on multiple drug use are extremely important. They suggest first that multiple drug use may be more prevalent in the general population than previously suspected. They also suggest the need for a new terminology and classification system for drug users. If “heavy” marijuana users are highly at risk for combined use of multiple substances, then attribution of effects/consequences based on classification of them according to the substance with which they are over involved may be inaccurate. The problems and troubles experienced by heavy marijuana users in particular may actually reflect interactions among the drugs being ingested rather than the impact of marijuana. Stated differently, perhaps it is time for attention to be directed toward overall consumption patterns in addition to queries about use of particular drug classes.

Treatment Outcome Prospective Study (TOPS)

If multiple drug use is found so readily in the general population, the prevalence of multiple use should be substantially higher among persons in treatment for drug abuse. Evidence from the 3389 drug abuse clients in the 1979 TOPS cohort confirms this expectation. The data in Table Weekly or Greater Use of Drugs by Primary Drug Problem: Treatment Outcome Prospective Study show the percentage of clients reporting weekly or more frequent use of various drugs by the “primary” drug with which they are having problems. For example, 1439 or 42.5% of the sample is classified with heroin as the primary problem. Among the primary heroin problem clients, the drugs and the percentage using them on a weekly or more frequent basis are: alcohol (53%); marijuana (64%); cocaine (42%); heroin (88%); illegal methadone (16%); minor tranquilizers (21%); and amphetamines (11%). While these figures are not from questions specifically about simulatneous combined use, the percentages reflect weekly or more frequent use of all of these substances by a group of heroin addicts in treatment across the United States. Multiple drug use is certainly the norm among those most heavily involved in the use of illegal drugs.

Table Weekly or Greater Use of Drugs by Primary Drug Problem: Treatment Outcome Prospective Study (TOPS)

Primary drug problem Drugs used weekly or more often (%)
N Alcohol Marijuana Inhalant Hallucinogen Cocaine Heroin
Alcohol 188 89.7 63.1 1.6 7.5 5.9 2.7
Marijuana 239 64.7 94.9 0.4 8.1 10.9 4.2
Inhalants 8 50.0 75.0 62.5 12.5 12.5 0.0
Hallucinogens 54 77.8 85.2 3.7 66.7 20.4 7.4
PCP” 18 100 94.4 0.0 44.4 16.7 0.0
Cocaine 125 68.6 71.5 3.2 9.8 81.5 27.2
Heroin 1439 52.6 64.2 0.4 1.5 42.2 87.8
Illegal methadone 163 50.6 55.0 0.0 1.9 17.3 20.9
Other narcotics 317 50.2 52.1 1.0 1.0 11.9 29.5
Minor tranquilizers * 88 54.1 59.8 0.0 6.9 9.2 8.0
Librium/Valium 8 62.5 37.5 0.0 0.0 12.5 12.5
Majortranquilizers 6 66.7 83.3 0.0 0.0 16.7 0.0
Barbiturates 39 76.9 73.7 2.6 13.2 23.1 10.3
Sedatives 25 72.0 70.8 4.0 16.7 8.3 4.4
Amphetamines 125 66.4 75.2 2.4 8.8 12.9 6.4
Preludin * 17 23.5 76.5 0.0 0.0 17.7 35.3
Other 24 66.7 54.2 4.2 8.3 20.8 25.0
No problem 442 51.4 54.4 1.4 3.5 6.0 8.9
Missing 64 66.7 75.0 9.7 19.7 15.0 11.5
All respondents 3389 57.4 64.9 1.2 5.2 26.9 45.1

* Data for clients mentioning a specific drug (phencyclidine-PCP, Librium/Valium, Preludin) as the primary problem are not included in the data for the more general drug types (hallucinogens, minor tranquilizers, amphetamines).

National Drug/Alcohol Collaborative Project (NDACP)

This study was conducted from 1974 through 1978 and involved 1544 clients in alcohol and drug treatment programs around the country. Those who were “regular” users of various substances were asked if they used another substance to “boost, balance, counteract, or sustain” the effects of that drug. If the response was yes, they were asked to specifically identify which substances were used for that purpose for each drug.

Listed below are the drugs included in the analyses and the percentage of regular users reporting use of other substances to enhance the effects of that drug.

Alcohol 53% Illegal methadone 72%
Heroin 71% Cocaine 77%
Other opiates 66% Hallucinogens 72%
Amphetamines 72% Inhalants 36%
Barbiturates 76% OTC drugs 54%
Minor tranquilizers 55% Antidepressants 80%
Marijuana/hashish 76% Major tranquilizers 57%

For alcohol and the hallucinogens, the major altering substance is marijuana. For heroin, cocaine and marijuana are the primary enhancement drugs, while heroin provides a buffering effect for cocaine among these clients. Heroin and alcohol are used to alter the effects of taking illegal methadone. For the remainder of the drugs, alcohol is the drug used most often to alter, enhance, or counteract the effects of the drug.

Preliminary Conclusions

There are at least two conclusions concerning multiple drug use that seem virtually beyond refutation. First, multiple drug use is not limited to those most heavily involved in the drug culture, although the heavier the involvement the more likely the occurrence of multiple drug use. The evidence for this conclusion is that multiple drug use seems to be part of a developmental process of drug involvement in the general population and seems to be rather prevalent in those segments of the population clearly identifiable because of their problems with drugs. Second, there is a relatively small proportion of the population that meets the most rigorous definition of multiple drug use (i.e., simultaneous use of multiple substances to either enhance or counteract the effects of the other drug). However, the simultaneous multiple drug users constitute a large enough percentage of the population to justify serious concern.

There are still a number of questions to be answered about multiple drug use from an epidemiological perspective. First, what is the extent of multiple drug use and has it increased over the past decade or so? Second, has the age at which onset of multiple drug use (i.e., particularly simultaneous use to enhance or counteract effects) increased, decreased, or remained the same? Third, what is the “normative” hiatus between using drugs separately for their individual effects and together for their multiple effects? Fourth, what are the predictors that differentiate those who become multiple users from those whose use is essentially exclusive? These are not new questions. They are the same questions that are raised whenever a “new” epidemic of drug use occurs in society. However, the questions take on added significance when there may be a multiplier effect present for consequences associated with the drugs being mixed on multiple use occasions. This last point is discussed in the next section.


Selections from the book: “Recent Developments in Alcoholism. Volume 4: Combined Alcohol and Drug Abuse. Typologies of Alcoholics. The Withdrawal Syndrome. Renal and Electrolyte Consequences.” Edited by Marc Galanter. An Official Publication of the American Medical Society on Alcoholism, the Research Society on Alcoholism, and the National Council on Alcoholism. 1986.