doses are required to yield a given response. The maximum effect of the stimulus is shown by the peak of the curve on the vertical axis. Finally, the steepness of the slope of the curve depicts the difference between ineffective and effective doses with respect to producing the effect of interest.

Dose-response curves are routinely used to study some properties of psychoactive drugs (Gable, 1993; Julien, 1998) —in particular, the effective dose (ED), which is usually the dose needed to produce the drug’s major psychopharmacological effect in 50 percent of respondents and the lethal dose, (LD), which is the dose sufficient to kill 50 percent of respondents. The ratio between the effective dose and the lethal dose is a common index of the drug’s overdose potential, and common street drugs vary widely in their ED:LD ratios. These analyses are largely based on laboratory analyses, often using animal subjects. Much less common are dose-response analyses based on field research, examining psychoactive drug use by humans in naturalistic field settings. In contrast, such analyses are fairly common in the epidemiological literature on alcohol consumption. For example, we have fairly good epidemiological evidence on the dose-response relationships between alcohol consumption and road fatalities, stroke fatalities, breast cancer, and violent criminality (Corrao et al., 1999; Edwards et al., 1994).

Very little is known about the dose-response relationships for the consequences of drug use depicted in Box 2.2. Understanding those dose-relationships is an ambitious, but in our view a fruitful, research agenda. Among the questions that might be addressed:

  1. What are the acute risks of a given incident of drug use? In other words, what is the dose-response relationship, for a given consequence, across the range of doses actually taken in street use of each drug?

  2. What are the chronic risks of a career of drug use? In other words, what is the dose-response relationship if the horizontal axis is converted to cumulative dose over time? Chronic risks will vary with the user’s experience with the drug (e.g., tolerance reduces overdose potential at any given dose, but the poor health consequences of prolonged use may leave experienced users more vulnerable in other ways). Cumulative risks are often difficult for casual users to observe; by definition, such evidence takes longer to accumulate, and heavy users are often socially isolated from casual users. Moreover, there is psychological evidence that people have difficulty appreciating that statistically small acute risks can accumulate to large probabilities with repeated exposure to a hazard (Doyle, 1997).

  3. How does the cumulative dose-response relationship over a drug-using career vary according to rate and tempo at which that total dose is accumulated? Some risks may vary mostly as a function of the quantity



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement