Many politicians, 'news' media, & entertainers have recurring claimed that heroin use is on the rise and that "glamorization" of heroin use by films such as Pulp Fiction have increased heroin use. Whether heroin use is glamorized by some media or not, the question remains "Are people using more heroin than they were in the past?".

The short answer is that the data is inconclusive, but it appears that heroin usage in the 1990's is higher than it was in the 1980s but perhaps similar to the 1970s. Heroin use has fluctuated down and up and down in the last 25 years. Some studies show that heroin use in the late 1990s was similar to the levels in the late 1970's. Other indicators show heroin use to be somewhat lower in 1998 than in the preceding several years and lower than the use in the late 1970s. The main thing when looking at data on drug use is that the numbers vary from year to year and the margin of error on the studies is actually quite high. Look for several year trends in use and also look for trends over 10 or 20 years to find meaningful changes.

Jennifer Stanhouse wrote an article in the fall 1996 American Sociological Association's Alcohol and Drugs Section Newsletter (Volume 6, Number 1, Fall 1996) titled "Heroin Is Staging A Comeback As A Drug of Choice, Or Is It?", in which she refutes the claims that the perceived "glamorization" has impacted heroin use. She points out that media-glamorization would be expected to increase youth usage, yet the age of heroin initiation (the age at which people first try a drug) increased until the last year of her data (1992), using Johnston, Gerstein, & Choy "Trends in the incidence of Drug Use in the United States, 1919-1992' (DHHS Publication # (SMA) 96-3076, Washington D.C: Superintendent of Documents, US Gov Printing Office).

The data from this report differs somewhat from the National Household Survey (data below), but the general trends appear the same. According to the NHSDA data, heroin initiation age has dropped in the last few years (most recent data from 1997) and the total number of initiations has increased from the levels in the early 1990s, but the data appears confused or possibly wrong because of its wild fluctuations from year to year.

According to the 1999 National Household Survey, the lifetime prevalence of heroin use in 1979 was 1.3% of the population, in 1998 it was 1.1%. The NHSDA study also estimates that the total number of people who had used heroin in the last year in the United States was 427,000 in 1979 and 253,000 in 1998, although their estimates for 1995-1997 are much higher (428,000 to 597,000 total heroin users). It appears that the NHDSA numbers do not provide a reliable trend in heroin use, but overall use has not gone up much since 1979 and may have gone significantly down.

From the Long Term Use statistics in the Monitoring the Future studies, we can see that the precentage of highschool seniors who report ever having tried heroin has fluctuated from between about 1% to about 2% over the last 25 years.

Monitoring The Future, % Highschool Seniors Ever Used Heroin
1970's
					1975	1976	1977	1978	1979
no data					2.2	1.8	1.8	1.6	1.1
1980's
1980	1981	1982	1983	1984	1985	1986	1987	1988	1989
1.1	1.1	1.2	1.2	1.3	1.2	1.1	1.2	1.1	1.3
1990's
1990	1991	1992	1993	1994	1995	1996	1997	1998	1999
1.3	0.9	1.2	1.1	1.2	1.6	1.8	2.1	2.0	2.0

National Household Survey's table on heroin initiation
			AGE-SPECIFIC RATE OF FIRST USE1
YEAR	INITIATES (1000s)	MEANAGE	12-17	18-25	26-34
1965	*	*	**	0.5	**
1966	*	*	**	0.4	**
1967	*	*	0.2	0.4	0.7
1968	27	19.4	0.3	0.5	0.3
1969	83	17.4	1.9	1.3	0.1
1970	97	19.3	0.3	3.0	0.1
1971	130	17.8	2.1	2.7	**
1972	140	17.7	1.4	3.3	0.1
1973	70	18.8	1.1	1.3	**
1974	81	22.8	0.3	1.6	0.9
1975	76	19.3	0.3	2.1	0.1
1976	79	19.8	0.9	1.1	0.7
1977	92	22.0	0.6	1.5	0.9
1978	69	20.3	0.1	1.7	0.3
1979	68	20.9	0.7	1.3	0.3
1980	55	20.8	0.3	0.9	0.3
1981	71	23.0	0.4	1.0	0.8
1982	44	21.4	0.1	0.9	0.2
1983	75	24.6	0.5	1.0	0.4
1984	86	26.9	0.3	0.7	1.0
1985	38	23.8	0.1	0.7	0.3
1986	63	20.9	0.9	1.1	0.2
1987	52	20.6	0.3	1.1	0.2
1988	72	27.4	0.2	1.0	0.2
1989	48	24.0	0.4	0.8	0.3
1990	66	26.4	0.3	0.7	0.7
1991	54	24.7	0.3	0.6	0.5
1992	41	23.0	0.4	0.6	0.1
1993	62	20.1	0.9	1.0	0.4
19942	85	21.2	1.2	1.3	0.5
19953	88	19.7	1.6	1.5	0.3
19964	149	18.3	2.7	2.6	0.2
19975	81	17.6	1.1	2.0	**

¹ The numerator of each rate equals the number of persons who first used the drug in the year (times 1000). The denominator of each rate equals the number of persons who were exposed to risk of first use during the year, weighted by their estimated exposure time measured in years. For example, for the age group 12-17 in 1990, the denominator is the sum of three components:

(1) those persons 12-17 years old in 1990 who first used the drug in 1989 or earlier, times a weight of zero. The weight is zero since they had zero exposure to the risk of first use in 1990.

(2) those who first used the drug in 1990 times a weight of .5. The weight of .5 assumes that these people, on average, first used the drug at midyear and consequently have a half year of exposure (i.e. the first half of the year.)

(3) those who never used, or those who first used the drug in 1991 or later, times a weight of one. The weight of one assumes their exposure to the risk of first use during 1990 was for the whole year.

Each person is also weighted by his/her sample weight.

² Estimated using 1995, 1996, 1997 and 1998 data only.

³ Estimated using 1996, 1997 and 1998 data only.

⁴ Estimated using 1997 and 1998 data only.

⁵ Estimated using 1998 data only.

Source: SAMHSA, Office of Applied Studies, National Household Survey on Drug Abuse, 1994-1998.