Researchers examined the impact of perceived substance function and negative substance-related experiences on drug use behavior in polydrug users. Participants were 364 polydrug users residing in or near London, England (205 men and 159 women, average age 19.3 ± 1.9), with participants selected on the basis of their reporting having used at least two of four drugs (cannabis, amphetamines, Ecstasy and cocaine) on at least 5 occasions within the 90 days prior to interview. None of the participants were receiving treatment for substancd abuse or dependence. Participants provided drug use histories and responded to instruments assessing substance-related adverse consequences and negative effects, perceived peer substance use, and perceived functions (e.g. changing mood, socializing) served by each substance used. In this study, the measure of adverse effects referred to unfavorable psychological (e.g. "preoccupation or worry"), social (e.g. "role neglect") and financial ("prioritizing spending money on substance use") consequences of use, and not medical adverse events. The measure of negative effects assessed out-of-control drug use ("taking more than you would have liked,") drug-use related accidents, and unwanted or unfavorable acute effects (such as anxiety). 177 of 364 participants (48.6%) reported having used Ecstasy at least once, and they reported first using Ecstasy at the age of 16.5 years (average age of use onset). 155 of 177 (85.7% of Ecstasy users, 42.6% of entire sample) reported using Ecstasy within 90 days of the interview. Average number of Ecstasy uses within 90 days was 9 times (range 1-51 times), and average Ecstasy dose per use was reported to be 1.7 tablets (range 0.5 to 5 tablets). Most respondents used alcohol and cannabis (327 reported alcohol use, and 350 reported cannabis use), and amphetamine and cocaine were used by a higher percentage of the sample than Ecstasy. Results indicate that Ecstasy use begins at a comparatively late age, after most people have sampled other substances, such as alcohol (13.1) or cannabis (14.4). Analyses of drug use parameters used "intensity of use" (frequency of use times average dose per use) because in most cases, the two variables were strongly related, and analyses of reported substance use functions considered only data for drug use in the last year. Most respondents indicated that they used Ecstasy either to enhance social events or to manage negative moods, a pattern of functions reported for other substances, such as alcohol and cannabis. The average number of negative effects associated with Ecstasy use was 2.99 (range 0-13), compared with 2.06 for cannabis (range 0-11) and 3.77 for amphetamine (range 0-13); the greatest number of negative effects were associated with alcohol (4.75, range 0-13). Perceived peer use of Ecstasy was not reported, but must be between 0.41, or two-fifths (for amphetamines) and 0.9 for alcohol. Regression analyses found that extent of peer use best predicted intensity of Ecstasy use in the past year, followed closely by high "managing negative mood" scores, Ecstasy use in best friend/partner, and age of first use, with younger onset of Ecstasy use associated with greater intensity. Number and extent of negative effects predicted intensity of Ecstasy use, but by a much smaller degree than the factors listed above. Regression analyses found that extent of problems ("adverse effects," or consequences) from Ecstasy use was best predicted by intensity of use, followed by high negative effects score, high scores on "managing negative mood" function, high scores on "enhance social events’ function, and (earlier) onset of use. As was true for other substances, perceived function of Ecstasy was a more powerful predictor of intensity of use and number of problems than was experiencing negative effects. These findings suggest that drug use behavior is multiply determined, with drug use guided more by perceived functions than by presence or absence of experiencing negative effects. It is notable that using Ecstasy for enhancement of social functions failed to predict intensity of use, possibly because scores on this measure did not vary much across participants (that is, there was not enough differences to create a relationship). It is also notable that people seeking to use substances (not just Ecstasy) for managing negative moods had the highest rates of drug-related problems, an association that may arise for a variety of reasons, including pre-existing causes of both the desire to manage negative moods and increased risk for drug-related problems. The researchers propose that drug abuse prevention programs should attend to the role and importance of perceived drug functions and should seek to teach people ways of dealing with negative moods that do not involve substance use. Findings also suggest that while somewhat important, the role of perceived peer use and experiencing negative drug effects do not strongly predict drug use behavior.