1997 National Household Survey on Drug Abuse:  Preliminary Results

Previous Page TOC Next Page


Estimates of drug use incidence, or initiation, provide another measure of the Nation's drug problem. They can suggest emerging patterns of use, particularly among young people. In the past, increases and decreases in incidence have usually been followed by corresponding changes in the prevalence of use. SAMHSA recently released a detailed report on incidence trends based on 1991-93 data, covering the period 1919 through 1992. Using the 1994 through 1997 NHSDA data, this section updates these earlier estimates and extends them to 1996.

Details of the methodology are available in Trends in the Incidence of Drug Use in the United States, 1919-1992, released in 1996. Briefly, the estimates are based on the NHSDA questions on age at first use. Using each respondent’s reported age at first use in conjunction with his/her age and interview date, the respondent’s year of first use of each drug was determined by subtracting their age from the interview year and then adding the age at first use. By combining all respondents and applying sample weights, estimates of the number of new users of each drug for each year were made. These estimates include new users at any age, including ages under 12. In addition, the average age of new users in each year and age-specific rates of first use were estimated. These rates are presented in this report as the number of new users per 1,000 person-years of exposure. The rates could also be described as the number of new users per 1,000 potential new users. The numerator of each rate is the number of persons in the age group who first used the drug in the year (times 1,000), while the denominator is the number of persons who were exposed to the risk of first use during the year, adjusted for their estimated exposure time (exposure time was expressed in years). Persons who first used the drug in a prior year have zero risk of first use in the current year, and persons who still have never used the drug by the end of the current year had one full year of exposure to risk. Persons who first used during the year are assumed to have a half year of exposure to risk.

The incidence estimates are based on retrospective reports of age at first drug use by survey respondents interviewed during 1994-97, and may therefore be subject to several biases, including bias due to differential mortality of users and nonusers of each drug, bias due to memory errors (recall decay and telescoping), and underreporting bias due to social acceptability and fear of disclosure. See Appendix 2, Section III for a discussion of these biases. As is explained in Appendix 2, it is possible that some of these biases, particularly telescoping and underreporting because of fear of disclosure, may be affecting estimates for the most recent years more significantly. However, analyses have not clearly shown the magnitude of these biases.


oAn estimated 2.5 million Americans used marijuana for the first time in 1996, about the same number as in 1995. The number had been increasing since 1991, after a long-term decrease that had been occurring since 1975. It is interesting to note that the decrease in prevalence of marijuana use that occurred in the 1980s did not begin to occur until several years after the peak in incidence.

oThe rising incidence during the 1990s seems to have been fueled primarily by the increasing rate of new use among youth age 12-17 years (from 37 per 1,000 person years in 1991, to about 83 per 1,000 person years in 1996). This is in contrast with the epidemic of the late 1960s and early 1970s, in which there were similar increases among young adults as well as youth, but with the rates among young adults dominating. The rate of marijuana initiation for youth in 1996 is similar to the estimated rates in the late 1970s, the peak years for marijuana incidence among youth (Figure 13).

Undisplayed Graphic


oThere were an estimated 171,000 new heroin users in 1996. Estimates of heroin incidence are subject to wide variability and usually do not show any clear trend, although there is a statistically significant upward trend in the number of new heroin users from 1992 to 1996, a finding that is consistent with anecdotal reports of increasing numbers of new heroin users. The recent increases in new heroin use are comparable to the increases seen in the epidemic of the late 1960s. The rate of heroin initiation for the age group 12-17 increased from below 1.0 during the 1980's to 3.9 in 1996 (Figure 14).

Undisplayed Graphic

oA large proportion of the recent heroin initiates are young and are smoking, sniffing, or snorting heroin. Among recent initiates found in the 1996 and 1997 NHSDAs, 90 percent were under age 26 and 75 percent had never injected heroin. A similar analysis of new heroin users in the 1991 and 1992 NHSDAs showed that only 61 percent were younger than age 26 and only 46 percent had never injected (questions about smoking, snorting, and sniffing were not included in the NHSDA until 1993).

Cocaine and Crack Cocaine

oThe annual number of new users of any form of cocaine rose from 1992 to 1996, but was at a lower level than during the early and mid-1980s. In 1996, there were an estimated 675,000 new cocaine users, compared with about 1.2 to 1.3 million cocaine initiates per year from 1980-1986.

Undisplayed Graphic

oThe rates of initiation among different age groups have been changing in recent years. In particular, the rate among youth age 12-17 increased from 4.0 in 1991 to 11.3 in 1996; this rate in 1996 was similar to the high rates of the early 1980s. Historically, most initiation of cocaine use has taken place among young adults age 18-25. The rate for that age group fell from a high of 29.8 in 1980 to 10.3 in 1992. Initiation rates among this age group have been fairly stable from 1991-1994, but started showing a trend toward increases since 1995; the rate in 1996 was 14.8. The rate of first use for ages 18-25 in 1996 is only slightly higher than the rate among the 12-17 age group (Figure 15).

oThe number of new crack cocaine users decreased from 382,000 in 1995 down to 248,000 in 1996. In particular, the rate of initiation of crack cocaine use among those age 18-25 year declined from 4.7 in 1995 to 3.2 in 1996.


oThere were an estimated 1.1 million new hallucinogen users in 1996, approximately twice the average annual number during the 1980s. The rate of initiation among youth age 12-17 increased between 1991 and 1996, from 11.7 to 25.8 per 1,000 person years. Over the same period, the rate for ages 18-25 years rose from 14.3 to 21.2, but the increase was not statistically significant. Both age groups had similar rates in 1995 and 1996.


oThere were an estimated 805,000 new inhalant users in 1996, up from 382,000 in 1991. The rate of first use among youths age 12-17 rose significantly from 1991 to 1996, from 10.3 to 21.0 per 1,000 person years. For young adults age 18-25, there was also an increase in the rate of first use between 1991 and 1996 (from 4.6 to 12.4 per 1,000 person years).


oAn estimated 3.3 million people tried their first cigarette in 1995 (1996 estimate not available). The rate of initiation among youth age 12-17 continued to increase from 1991 to 1995.

oAn estimated 1.9 million people began smoking on a daily basis in 1996, and the rate of youth initiation of daily smoking increased somewhat from 1991 to 1996. The average age at which people started smoking daily has been declining somewhat since 1993, although the change is not statistically significant.


oIn 1995 there were approximately 4.3 million new users of alcohol, compared with about 3.5 million new users in 1991. The rate of new usage of alcohol among the 18-25 age group has begun to rise in recent years (168 per 1,000 person years in 1989, compared with 243 per 1,000 person years in 1995). In addition, the rate of initiation of alcohol use among the 12-17 age group increased from 113 per 1,000 person years to 165 over the same period.

Previous Page Page Top TOC Next Page

This page was last updated on February 05, 2009.