Trends in Drug Use - Trends In Incidence

Before people begin to use a drug more or less regularly, they have to use it for the first time. The government's drug experts call first use of a drug its "incidence" of use or the event of "initiation." The government's chief drug survey, the National Survey on Drug Use and Health (formerly called the National Household Survey on Drug Abuse, conducted annually by the Substance Abuse and Mental Health Services Administration (SAMHSA) of the U.S. Department of Health and Human Services, tracks both first use of important drugs and their prevalence. Prevalence, discussed in the next section, is the extent of current and lifetime use of drugs by the population. Increases in incidence have been found to foreshadow increases in prevalence; similarly, when the number of initial uses of a drug drop, after a lag of years so will the number of people who regularly take the drug.

The survey began in 1971 and has increased from a survey of about three thousand respondents every two to three years to almost seventy thousand people in the fifty states and Washington, D.C., every year. The latest survey available, 2003 National Survey on Drug Use and Health (NSDUH), was published in September of 2004 (http://www.oas.samhsa.gov/nhsda/2k3nsduh/2k3ResultsW.pdf). The population surveyed by SAMHSA consists of noninstitutionalized civilians over the age of twelve living in households, dormitories, homeless shelters, rooming houses, and military institutions. This excludes homeless people not in shelters, active-duty military personnel, and persons in jails and prisons, but the survey is still considered to be the most comprehensive analysis of drug use in America. The results are statistically projected to the entire population to produce an estimate of drug use prevalence nationwide. Survey methods were changed in 1999 when SAMHSA switched from a paper-and-pencil survey to a computer-assisted survey. Therefore, the results for the years from 1999 forward are not strictly comparable to earlier years. The changes were introduced so that more accurate state-level results could be obtained and over-or under-sampling of regions (urban versus rural, for instance) or populations (African-Americans versus whites) could be corrected.

SAMHSA tracks initial use by asking those who participate in the National Survey when they first used a drug. Respondents also report how old they are. SAMHSA can thus calculate the number of people first using a drug in any given year—and also how old they were at that time. People who use drugs have a higher rate of mortality than nonusers; current samples cannot, of course, include the dead. Reporting on the "harder" drugs is also less reliable because of what SAMHSA calls "underreporting due to desire for social acceptability or fear of disclosure" ("Chapter 5. Trends in Initiation of Substance Use," 2003 National Survey on Drug Use and Health [NSDUH], Rockville, MD: SAMHSA, 2004).

Initial use shows different cycles for different drugs over the last four decades. The peaks and troughs of different drugs do not always coincide. The patterns suggest a demographic underpinning, since initial drug use is generally a youth phenomenon, and different drugs are used more by some age groups than others. As the number of people in each age category shifts, so do the incidents of drug usage in each category.

Marijuana

By far the most frequently tried illegal substance is marijuana. In the thirty-seven-year period shown in Figure 3.1, initial use of marijuana shows several peaks, with two of the highest in 1973 and 2000. In 1973, 3.5 million people tried marijuana for the first time. Initial tries dropped to about one and a half million by 1990. New tries then climbed to a new peak of 2.9 million first-time users in 2000, before tapering off slightly to 2.6 million in 2002, about two-thirds of whom were under age eighteen.

FIGURE 3.1
New users of marijuana, 1965-2002
SOURCE: "Figure 5.1. Annual Numbers of New Users of Marijuana: 1965-2002," in Results from the 2003 National Survey on Drug Use and Health: National Findings, U.S. Department of Health and Human Services, Substance Abuse and Mental Health Services Administration, Office of Applied Studies, 2004, http://www.oas.samhsa.gov/nhsda/2k3nsduh/2k3ResultsW.pdf (accessed February 10, 2005)

The demographic underpinning is suggested by the age of members of the baby boom generation at the 1973 peak of marijuana experimentation and the age of the so-called "baby boom echo" (children of the baby boom) in 2000. A large percentage of the boomers (born between 1945 and 1960) were between sixteen and twenty-five years of age in the mid-1970s; at around forty million strong, this was the largest single group of this age ever in America ("Live Births by Age of Mother and Race: United States, 1933-98," Washington, DC: National Center for Health Statistics, http://www.cdc.gov/nchs/data/natality/mage33tr.pdf).

The second peak in this series (in 2000) suggests that the children of those who experimented in the 1970s were now doing the same thing. In 2000 the youngest boomers were forty and the oldest fifty-five, and all were of an age to have teenagers in the house or away in college.

With the passage of time during this thirty-seven-year period, the younger age group also became more important. Experimentation is driven by the under-eighteen age group. (See Figure 3.1.) SAMHSA also calculates the average age of the initial users. The general trend over time has been that those trying marijuana are younger and younger on average. The average age in the late 1960s was about nineteen; the average age in 2002 was 17.2.

Cocaine

Data on the incidence of cocaine use produce a different pattern in the 1965-2002 period. (See Table 3.1.) The number of initial users reached its peak in 1980 (1.56 million), seven years after the first marijuana peak.

TABLE 3.1
Number of new users of cocaine, 1965-2002
[In thousands]
SOURCE: "Table G.32. Numbers (in Thousands) of Persons Who First Used Cocaine in the United States, Their Mean Age at First Use, and Rates of First Use (Per 1,000 Person-Years of Exposure): 1965-2002, Based on 2002 and 2003 NSDUHs," in Results from the 2003 National Survey on Drug Use and Health: National Findings, U.S. Department of Health and Human Services, Substance Abuse and Mental Health Services Administration, Office of Applied Studies, September 2004, http://www.oas.samhsa.gov/nhsda/2k3nsduh/2k3ResultsW.pdf (accessed February 10, 2005)

	Number of initiate (1,000s)				Age specific rates^a
Year	All ages	Under 18	18 or older	Mean age	12-17	18-25
1965	62	*	49	*	*	*
1966	35	*	27	*	*	*
1967	98	*	98	*	*	*
1968	185	67	118	19.2	2.9	5.0
1969	210	50	160	21.6	2.0	5.0
1970	288	85	204	18.9	3.7	8.0
1971	424	81	343	20.7	3.3	11.0
1972	433	111	322	19.5	4.5	10.8
1973	592	174	418	20.3	7.2	13.5
1974	885	202	683	21.7	8.3	18.1
1975	876	197	679	20.9	8.1	19.6
1976	1,026	212	813	21.2	8.9	24.3
1977	1,102	264	838	21.2	11.3	23.4
1978	1,329	237	1,092	21.3	10.4	31.7
1979	1,360	277	1,083	21.0	12.4	32.9
1980	1,562	296	1,266	22.4	13.6	33.0
1981	1,543	284	1,259	21.5	13.5	36.3
1982	1,502	299	1,203	22.0	14.3	32.7
1983	1,450	235	1,215	22.1	11.4	35.5
1984	1,363	267	1,097	22.1	13.1	30.8
1985	1,371	289	1,082	22.2	14.4	29.6
1986	1,264	231	1,033	23.2	11.7	28.3
1987	1,173	247	926	22.5	12.6	26.8
1988	1,147	166	982	23.1	8.6	28.3
1989	981	163	818	23.1	8.5	22.8
1990	942	195	748	23.5	10.4	19.7
1991	706	115	591	23.1	6.0	16.6
1992	781	166	615	24.5	8.3	14.3
1993	634	150	484	22.1	7.5	14.3
1994	662	151	511	22.6	6.9	14.4
1995	776	222	554	21.1	10.3	17.2
1996	880	272	608	21.9	12.3	18.4
1997	886	252	634	20.8	11.1	21.3
1998	926	278	648	20.8	12.0	20.9
1999	959	283	677	20.8	12.0	21.5
2000	1,019	309	709	20.9	12.8	22.4
2001	1,167	372	795	20.8	15.3	24.1
2002^b	1,059	323	736	20.3	13.1	22.6
^*Low precision; no estimate reported.
Note: Comparisons between years, particularly between recent estimates and those from 10 or more years prior, should be made with caution due to potential reporting and other biases.
^aThe numerator of each rate is the number of persons in the age group who first used the drug in the year, while the denominator is the person-time exposure of persons in the age group measured in thousands of years.
^bEstimated using 2003 data only.

Cocaine is used by an older age group. The average age of cocaine initiates was around twenty-two years in the early 1980s, rose to about twenty-three in the late 1980s, then declined to about twenty-one by the late 1990s. The average age of initiates in 2002 was 20.3 years.

Until the mid-1980s cocaine was a relatively expensive powdered drug snorted by well-off users at parties. The much cheaper crack cocaine, which could be smoked, appeared early in the 1980s but did not reach mass distribution until some years later.

Hallucinogens, Inhalants, and Prescription Drugs

Demographics underlie but do not entirely explain the incidence of drug use. The availability of drugs, their cost, the emergence of new varieties, the dangers associated with the drug (and the spread of information about such dangers) all have a bearing. Hallucinogens, inhalants, and psychotropic medications used in nonmedical settings have similar usage patterns.

HALLUCINOGENS.

Drugs that produce hallucinations reached their first-use peak in 2001. (See Figure 3.2.) The best known and most commonly used of these drugs is LSD (as measured in prevalence). The two other hallucinogens tracked by SAMHSA are PCP and Ecstasy (MDMA). In 2001, according to SAMHSA's 2003 report, 1.6 million people first took a hallucinogenic drug. The earlier peak in usage came in 1972, around the same time marijuana reached its highest incidence.

INHALANTS.

Substances intended for other purposes but inhaled for an effect include glue, gasoline, paint, and turpentine. Inhalant first use increased during the 1990s, with teenagers generally fueling the trend. Though a fairly steady group of eighteen- to twenty-year-olds has continued to join the inhalers over the past four decades, the growth in incidents has been driven mainly by the younger grouping of those twelve to seventeen. More than half (52%) of those who sniffed such items in 1978 were twelve to seventeen. According to the SAMHSA's 2003 report, in 2002 about one million people tried inhalants for the first time. As in the past, this group was dominated by those under eighteen, representing 78% of new users.

PRESCRIPTION DRUGS.

SAMHSA data show that use of prescription-type pain relievers as recreational drugs rather than for medical purposes has slowly risen from 1965 to 1995, after which it dramatically rose. Similar trends are seen for other prescription drugs, such as tranquilizers, stimulants, and sedatives. Incidence of these drugs reached early peaks in the mid- to late 1970s. All but the sedatives reached new highs since 2000, with pain relievers leading the way. The average age of those using pain relievers was twenty-two in 2002, according to SAMHSA's 2003 report. About 2.5 million new people tried prescription pain relievers each year from 2000 to 2002.

Incidence of Heroin Use

People who try heroin for the first time are, on average, the oldest "drug experimenters." According to a previous SAMHSA report, in the 1965-2000 period the average age of initial heroin users was 22.8 years. In the 1980-2000 period the average was 24.8, and in the 1990-2000 period it was 25.4. The users also represent the FIGURE 3.2
New users of hallucinogens, 1965-2002
SOURCE: "Figure 5.2. Annual Numbers of New Users of Ecstasy, LSD, and PCP: 1965-2002," in Results from the 2003 National Survey on Drug Use and Health: National Findings, U.S. Department of Health and Human Services, Substance Abuse and Mental Health Services Administration, Office of Applied Studies, 2004, http://www.oas.samhsa.gov/nhsda/2k3nsduh/2k3ResultsW.pdf (accessed February 10, 2005) smallest number of people who try a drug. The 2003 SAMHSA study reports that from 1995 to 2002 the annual number of new heroin users ranged from 121,000 to 164,000. Incidence reporting on heroin may be the least reliable because of social bias associated with heroin addicts and because many addicts die and thus do not participate in SAMHSA's household surveys that look back on the 1960s and 1970s.