Given our lack of ability to definitively describe the extant population of injection drug users, it should be clear that any attempts to project into the future are highly speculative. Yet drawing some sense of factors that may influence the size of the injection drug user population in the coming years seems warranted. In referring to CEWG, trend analysis would seem to hold some promise as a basis of forecasting; however, many of the secondary indicator databases relating to substance abuse do not currently report on route of drug administration. As a consequence, CEWG does not currently monitor trends in injection drug use as an independent focus of analysis. Recent proceedings nevertheless include discussion of at least two trends, previously noted by Frank and Williams (1994), that may significantly influence future injection drug user prevalence. The first is AIDS-related mortality, which has already begun to deplete the numbers in this population. The second, and potentially countervailing trend, is a resurgence in the prevalence of heroin use, which has been associated with an increasing number of initiates who are using the drug intranasally.
The proceedings of the CEWG meeting in December 1993 show the trend of increasing intranasal heroin use to be most pronounced in Newark, Chicago, and New York City (National Institute on Drug Abuse, 1994a). Newark reported 66 percent of heroin admissions to be snorters (French and Mammo, 1994). In Chicago, intranasal consumption accounted for 60 percent of treatment admissions for heroin dependence (Wiebel et al., 1994), and in New York City, 51 percent of heroin admissions reported a primary intranasal route of drug administration (Frank and Galea, 1994). DEA's Domestic Monitor Program (Drug Enforcement Administration, 1994), which analyzes the price and purity of heroin in major metropolitan areas, has documented substantial increases in the purity of heroin sold on the streets over the past half decade. Some have noted a direct association between the availability of higher-quality heroin and its intranasal consumption (Ouellet et al., 1993; Des Jarlais et al., 1994; Friedman et al., 1994; Frank and Williams, 1994); however, it is not yet clear whether this regionally emergent epidemic will spread across the entire country. It is also uncertain as to the proportion of current intranasal heroin users who are or will become physically dependent, or the numbers likely to ultimately progress to injecting the drug. Ouellet et al. (1993) in Chicago noted that current younger snorters of heroin do not see themselves as being at any risk for progressing to injection, unlike older injectors, who consider intranasal heroin use as merely a transitional phase leading up to injection. As cautioned by Friedman et al. (1994) in a commissioned paper for this panel on the etiology of drug injection, "Heroin snorters represent a real but unknown risk for progressing to injection drug use."