Crime and Drugs

views updated


Because of widespread public and political concern over drug-related crime, there has been an urgent need to understand the drugs-crime relationship. However, despite numerous studies on this topic, only recently have significant empirical advances in our understanding emerged.

Authors of a comprehensive literature review published in 1980 concluded that the drugs-crime relationship was far more complex than originally believed (Gandossy et al.). While acknowledging significant contributions of previous research, the authors felt that methodological problems in the studies they reviewed had obscured an understanding of the linkage between drugs and crime. As these and other reviewers have observed, perhaps the most serious of these weaknesses was the use of official arrest records as indicators of criminal activity. Studies using confidential self-report methods in settings in which there is immunity from prosecution have consistently documented that less than 1 percent of offenses committed by drug abusers result in arrest.

With a subsequent emphasis on confidential self-report data, studies conducted since 1980 have permitted more realistic estimates of the extent of criminality among drug abusers. In addition, victims of violent crime are now being asked whether they perceived the offender to be under the influence of drugs or alcohol. The annual Bureau of Justice Statistics (BJS) National Crime Victimization Survey asks this question of crime victims. Though a subjective inquiry, the 1998 survey revealed that 30 percent of victims could not determine whether the offender was under the influence of a substance. Of those who could make a determination, about 31 percent reported that the offender was under the influence of drugs or alchohol.

In 1997, the National Institute of Justice (NIJ) established the Arrestee Drug Abuse Monitoring (ADAM) Program to measure drug use among arrestees by calculating the percentage of arrestees with positive urine tests for drug use. ADAM data are collected voluntarily and anonymously at the time of arrest in booking facilities in thirty-five U.S. cities. The ADAM program has given researchers a powerful tool for obtaining empirical evidence of patterns of drug abuse. ADAM is the only national research program studying drug use that employs both drug testing and interviews, giving analysts the means of assessing validity of self-report data. Therefore, ADAM data are less susceptible either to exaggeration or denial of drug use than many other surveys. Moreover, ADAM is the only national drug research program built upon data collection at the local level. This data has revealed that there is no single national drug problem, but rather different local drug problems that vary from city to city.


In examining the criminality of drug abusers, it is important to note that the onset of illicit drug use typically does not result in the onset of criminal behavior. Rather, it is the frequency, not the onset, of drug use that increases criminal activity. Further, the positive relationship between drug-use frequency and crime frequency is not consistent across all types of drug use and all types of crime. Such a relationship has been observed with respect to only three types of drug abuse: heroin addiction, cocaine abuse, and multiple-drug use. In addition, such associations are more common for property crime than for violent crime.

Narcotic Drug Use.

Much of our knowledge about the relationship between drugs and crime comes from detailed self-report information on the type, extent, and severity of criminal activity of Narcotic (mainly heroin) addicts. Large-scale, independently conducted studies have convincingly shown that increases in property crime and robbery, which has components of both property crime and violence, are associated with increased heroin use. Such a relationship, however, is less clear for violent crimes other than robbery.

Prevalence and Scope of Property and Violent Crime. Several key studies reveal an exceptionally high prevalence of property crime among narcotic addicts. Anglin and Speckart (1988) found that 82 percent of a sample of 386 California male narcotic addicts reported involvement in property crime over an average five-year period of daily narcotic use. Anglin and Hser (1987) reported that 77 percent of a sample of 196 female narcotic addicts from California admitted to involvement in property crime during an average six-year narcotic addiction period. Inciardi (1986) noted that almost all of a sample of 573 male and female narcotic abusers from Miami had reportedly engaged in theft during the year prior to interview. Inciardi also found that these individuals reported involvement in more than 77,000 property crimes (on average, 135 per subject) over a 12-month period while at large in the community. This figure included 6,669 burglaries, 841 vehicle thefts, 25,045 instances of shoplifting, and 17,240 instances of fencing. While these studies varied in sampling methods and definitions of property crime (e.g., including and not including robbery), all provided evidence that a substantial majority of narcotic abusers routinely engage in property crime.

Property crime comprises a considerable portion of the crime, other than drug distribution, committed by narcotic addicts. For instance, Nurco et al. (1991a) found that of the nondistribution crimes committed by a sample of 250 male narcotic addicts during an average 7.5-year addiction period, approximately 48 percent were property crimes.

Research has also consistently documented that among heroin addicts, violent crime is less prevalent and occurs with less frequency than property crime. Earlier studies noted that addicts tended to prefer property crime over violent crime and appeared to be less violent than other offenders. While findings from later studies have continued to show that violence accounts for only a small proportion of all addict crime (approximately 1% to 3%, a rate that is much smaller than the property-crime figure), the actual number of violent crimes is still quite large because addicts commit so many crimes. For example, in Inciardi's 1986 sample of 573 Miami narcotic abusers, violent crime comprised only 2.8 percent of all offenses (5% of non-distribution offenses) committed by the subjects in the year prior to interview. However, this relatively small percentage amounted to 6,000 incidents of violent crime (on average, 10.4 per subject), since a total of 215,105 offenses were committed.

Researchers have also suggested that heavy heroin use and, more recently, heavy cocaine abuse have contributed to record numbers of homicides in large cities in the United States. The ways in which drugs can contribute to violence is the basis for a prominent theory in the drugs-crime field, discussed later in this article.

Crime and Frequency of Heroin Use. Recent studies have provided consistent evidence of a direct, functional relationship between the frequency of narcotic drug (primarily heroin) use and the frequency of property crime. These investigations have employed a unique longitudinal design in which crime data are obtained for each subject over periods during which the frequency of narcotic use may vary. These studies of addiction careers reveal that property-crime rates are significantly higher during narcotic addiction periods than during periods of non-addiction. Such a relationship tends to be linear, with the highest property-crime rates occurring at the highest levels of narcotic use (three or more times per day). In addition, although most addicts commit property crime prior to addiction, the frequency of such crime increases significantly from preaddiction to addiction, remaining high over subsequent addiction periods and low during intervening nonaddiction periods. While other factors also influence property-crime rates, the simplest explanation for these results is that property crime is functionally related to narcotic addictionsince addicts need cash to support their habits.

Evidence of a similar relationship between heroin use and violent crime is less conclusive. Studies have consistently shown that rates for robbery, in which there are property-crime features, are considerably higher during addiction periods than during either preaddiction or nonaddiction periods. However, when rates for composite measures of violence and rates of assault alone are examined, the relationship appears less clear.

In compiling composite measures of violence, Ball et al. (1983) found that for a sample of 243 male Baltimore addicts, the number of days on which violent crime was committed was considerably higher during the first addiction period than during the first nonaddiction period. However, in subsequent studies of 250 male addicts from Baltimore and New York City, most of whom had multiple periods of addiction, more complex relationships were observed. Over an addiction career, violent-crime rates for the total sample were significantly higher for combined addiction periods than for combined nonaddiction periods (Nurco et al., 1986; Nurco et al., 1988a). This result stemmed largely from high levels of crime committed during the first addiction period; violent crime actually decreased over subsequent periods of addiction, a finding that appeared to be age-related. The fact that mean rates for violence were found to be even higher for preaddiction (10 days per year) than for addiction periods (8 days per year) also reflected an inverse relationship between age and violent criminal activity.

The 1999 ADAM report on U.S. drug use of arrestees reveals that opiate use among adult arrestees remains relatively low compared to the prevalence of cocaine and marijuana in the overall sample. For female arrestees the median rate for testing positive to opiates was 8 percent in 1999 and for male arrestees it was 6 per cent.

Nonnarcotic Drug Use.

Investigation of the nonnarcotic drugs-crime relationship has only recently emerged as a major research question. In the 1980 literature review, Gandossy and associates found that, of the few studies conducted on the nonnarcotic drug-crime relationship, evidence linking the use of various nonnarcotic substances to either property crime or violent crime was weak. Another reason for the unclear relationship between nonnarcotic drug use and criminal behavior is that various narcotic and nonnarcotic drugs are often used in combination. Thus, disentangling their separate relationships to criminal activity, let alone determining cause and effect, is especially problematic. Despite these difficulties, significant advances have been made in understanding the nonnarcotic drugs-crime relationship since 1980. Cocaine. Data analyses on a nationwide probability sample of 1,725 adolescents strongly supported a cocaine-crime connection (Johnson et al., 1991). Adolescents who reported using cocaine in the year preceding the interview (comprising only 1.3% of the sample) were responsible for a disproportionately large share of the property and violent crime committed by the sample during this period. The cocaine users accounted for 60 percent of all minor thefts, 57 percent of felony thefts, 41 percent of all robberies, and 28 percent of felony assaults committed by the entire sample.

Typological studies involving seriously delinquent youth and female crack-cocaine abusers also revealed that subjects who reported the heaviest levels of cocaine use also had substantially higher rates of property and violent crime than subjects who used crack less frequently. Among a sample of 254 youth identified by Inciardi and coworkers (1993a) as serious delinquents, the 184 Crack dealers (86% of whom were daily crack users) were responsible for 45,563 property crimes (an average of 231 per user) during the year preceding the interview. In contrast, the seventy subjects who were not crack dealers and who used crack less frequently (approximately three times per week) averaged 135 property crimes per year. In addition, the heavy cocaine users averaged ten robberies per year, compared with one per year for the remaining subjects. Similar results were reported for a sample of 197 female crack abusers (Inciardi et al., 1993b). The average adjusted annual rates for the 58 subjects classified as heavy cocaine users (8 or more doses per day) were 12, 14, and 320 for violent crime, major property offenses, and minor property crimes, respectively. These rates were substantially higher than rates for the 90 subjects classified as "typical" users (4-7.99 doses per day). For those forty-nine users who took less than four doses per day, the average adjusted annual rates for violence and major property crime were less than one, and the rate for minor property offenses was twenty-four.

Increased cocaine use among narcotic addicts has also been associated with increased property and violent-crime rates. Both Nurco et al. (1988b) and Shaffer et al. (1985) found that male narcotic addicts who had higher rates of cocaine use tended to have higher rates of property and violent crime than addicts who did not abuse cocaine.

The 1999 ADAM report found that cocaine use among adult arrestees remained high, with cocaine found in more than one-third of adult arrestees in twenty sites. There was substantial variation in the proportion of those testing positive for cocaine. In three sites (Atlanta, Chicago, New York City), more than 60 percent of adult female arrestees tested positive for cocaine. In six other sites, however, cocaine use was less than 25 percent.

Other Nonnarcotic Drugs. The use of other non-narcotic drugs appears to be unrelated to increased criminal activity. While there is considerable evidence that frequent users of multiple nonnarcotic substances, including amphetamines, Barbitu-Rates, marijuana, and PCP, typically have high crime rates (although somewhat lower than rates for heroin addicts), such is not the case for users of single non-narcotic drugs. Although single use may be related to offenses like disorderly conduct or driving while impaired, it is not generally associated with predatory crime.

Marijuana. Research on the relationship between marijuana use and crime has found that, with the possible exception of the sale of the drug and disorderly conduct or driving while impaired, the use of marijuana is not associated with an increase in crime. Some studies have reported that marijuana use may actually reduce inclinations toward violent crime.

A major problem in studying the association between marijuana use and criminal behavior is that the exclusive use of marijuana is generally short-lived. Further, like other illicit nonnarcotic substances, marijuana is often used in combination with other drugs. Under such circumstances, it is difficult to isolate the effects of heavy marijuana use from those associated with the use of various drug combinations.

The 1999 ADAM report disclosed that marijuana remains a very popular drug for adult arrestees, particularly among young males between 15 and 20 years old. The median rate of marijuana positives for this group of arrestees was 63 percent compared to the overall adult male arrestee median rate of 39 percent and the overall adult female arrestee median rate of 26 percent.

Amphetamines. Literature reviews published during the late 1970s and early 1980s (Gandossy et al., 1980; Greenberg, 1976) reported that the association between amphetamine use and crime was difficult to determine because, among other factors, of the diversity of amphetamine users. More recent ethnographic studies of drug abusers (Goldstein, 1986) have reported that amphetamine use is related to violent crime in some individuals. In the general population, however, the association between amphetamine use and crime is not readily apparent. Despite assertions of the media in the 1960s and 1970s, the prevalence of amphetamine-related violence among American youth is likely to be quite low.

The 1999 ADAM report indicated that methamphetamine use among ADAM arrestees is concentrated mainly in the Western part of the United States. A large number of sites had virtually no presence of methamphetamine. However, prevalence rates exceeded 10 percent both for adult female arrestees in twelve sites and for adult male arrestees in 9 sites.

Psychedelic Substances. Most studies investigating the relationship between psychedelic-substance abuse and crime have involved Phencyclidine (PCP). Much of this research has examined the relationship between PCP and violence. As in the coverage of many other nonnarcotic drugs, media reports, principally in the 1970s and early 1980s, emphasized a perceived link between PCP use and violent behavior. However, the actual extent of this link is greatly exaggerated. In his report on the subject, Kinlock (1991) noted that serious methodological problems in some studies and contradictory findings in others disallowed a conclusive answer to the question of whether PCP use increased violent crime. Researchers have suggested, nevertheless, that the inconsistency of study findings may indicate that PCP use facilitates violence in a small proportion of users (Inciardi, 1986; Kinlock, 1991). There is agreement that biological, psychological, situational, and other factors underlying seemingly drug-related aggressive behavior should be examined in future research.


Inciardi (1986) has noted that numerous theories have been posited to explain the drugs-crime relationship. Many of these theories have dealt with the etiology of drug use and crime. Early etiological theories tended to be overly simplistic, focusing on what Inciardi termed the "chicken-egg" question: Which came first, drugs or crime? This question polarized the drugs-crime field for over fifty years. It typically reflected two mutually exclusive positions: that addicts were criminals to begin with, and addiction was simply another manifestation of a deviant lifestyle; or that addicts were not criminals but, rather, were forced into committing crime to support their drug habits.

Reflecting a middle-ground position, more recent theories argue for a diversity among narcotic addicts with regard to the predispositional characteristics and motives underlying drug-related criminal behavior. For example, on the basis of their research with narcotic addicts, Nurco and his associates (1991b) concluded that there is considerable variation among addicts in their propensity toward criminal activity. Some addicts had been heavily involved in crime prior to addiction, whereas others were extensively involved in crime only when addicted.

In the late 1970s, drugs-crime theories became increasingly complex, partly because studies tended to have fewer methodological problems that interfered with the measurement of both drug use and crime. With improvements in techniques, researchers gradually become more aware of heterogeneity among drug abusers on many dimensions, including the type and severity of drug-use patterns and related criminal activity. Also, more recent studies have found that drug use and crime, in most instances, do not initially have a causal relationship but, rather, are often the joint result of multiple influences. Among the many factors contributing to drug use and/or crime are those involving the family, such as lack of parental supervision, parental rejection, family conflict, lack of discipline, and parental deviance; association with deviant peers; school dropout, failure, and discipline problems; and early antisocial behavior. Consistent with the notion that all drug abusers are not alike, varying combinations of factors probably contribute to different patterns of deviant behavior in individuals at risk.

However, as Inciardi and his associates (1993a) have noted, some limitations of these theories remain. Most theories discuss drug abuse only as one of several manifestations of delinquency. Further, as in earlier years, the primary concern has been with the etiology of deviant behavior. Very little attention has been paid to explaining events that occur after the onset of drug use and criminal behavior, specifically how certain types of drug abuse increase the frequency of criminal activity. Finally, theories have typically focused on adolescence, without incorporating attributes and events that influence behavior during childhood and adulthood.

Among the most prominent theories in the drugs-crime field is that of Paul Goldstein (1986, 1989) regarding the relationship between drugs and violence. Goldstein's theory is based on his numerous ethnographic accounts of violent drug-related acts obtained from both perpetrators and victims in New York City. According to this theory, drugs and violence can be related in three separate ways: psychopharmacologically, economic-compulsively, and systemically. Within the psycho-pharmacological model, violent crime results from the short- or long-term effects of the ingestion of particular substances, most notably crack-cocaine and heroin. According to the economic-compulsive model, violent crime is committed as a means to obtain money to purchase drugs, primarily expensive addictive drugs such as heroin and cocaine. The systemic model posits that drug-related violence results from the traditionally aggressive patterns of interaction found at various levels within systems of illicit-drug distribution. Examples include killing or assaulting someone for failure to pay debts; for selling "bad," or adulterated, drugs; or for transgression on one's drug-dealing "turf."

Several key studies have analyzed data in the light of Goldstein's concepts. In a study of 578 homicides in Manhattan in 1981, 38 percent of the male and 14 percent of the female victims were murdered as a result of drug-related activity (Tardiff et al., 1986). The investigators contended that these percentages were higher than those previously reported in the United States. In a subsequent study by Goldstein and his coworkers (1989) involving 414 homicides in New York City that occurred over an eight-month period, 53 percent were classified by the police and researchers as being drug-related. In both studies, most of the drug-related homicides were attributed to systemic causes. Interestingly, in the former study, most of the homicides involved heroin, whereas in the latter study, most involved crack-cocaine.

Drug Use and High-Rate, Serious Criminality.

As indicated earlier, the onset of illicit drug use typically does not result in the onset of criminal behavior. In most cases, both drug use and crime begin in the early teens. Generally, the less serious the drug or crime, the earlier the age at onset of involvement. For example, among illicit drugs, marijuana is more commonly used at a younger age than are sedatives or tranquilizers, and these drugs, in turn, are typically used at a younger age than are "hard" drugs, such as heroin and cocaine. Similarly, minor forms of crime (e.g., shoplifting, vandalism) have an earlier onset than more serious types of crimes, such as assault, robbery, and drug dealing.

Most marijuana users do not become heroin addicts, and most youths who commit minor property crimes do not subsequently become involved in more serious offenses. In both instances, the salient variable appears to be age of onsetthe younger the individual is when first using a "soft" drug or committing a minor crime, the more likely he or she will move on to more serious forms of deviance. In general, the more deviant the environment (family, peers, community), the earlier the onset of deviance.

Since 1980, independent studies have identified several core characteristics of high-rate, serious offenders. According to Chaiken and Chaiken (1990), these studies have consistently found that predatory individuals tend to commit many different types of crime, including violent crime, at high rates, and to abuse many types of drugs, including heroin and cocaine. Also, research findings have consistently reported that among heroin addicts, prisoners, and seriously delinquent youth, the younger one is at onset of heroin and/or cocaine addiction, the more frequent, persistent, and severe one's criminal activity tends to be. In these studies, individuals with early onsets of addiction (typically before age 16) tended to abuse several types of drugs and to have disproportionately high rates of several types of crime, regardless of addiction status. Such findings have been observed in various geographic locations and are independent of ethnic group. These results are similar for both males and females, with one notable exception: females with early onsets of addiction are more likely to commit prostitution, shoplifting, and other property crimes at high rates, whereas males with early onsets are more likely to commit violent acts.

Chaiken and Chaiken's 1982 study of over two thousand male prisoners in three states was significant for at least two reasons. First, it challenged the long-held perception that drug abusers were less violent than other arrestees. While 65 percent of Chaiken and Chaiken's sample reported having used illicit drugs during the one- to two-year period preceding the arrest leading to the most recent incarceration, an even higher proportion (83%) of high-rate, serious offenders, identified as "violent predators," had used drugs during the same period. Among the offenders studied, violent predators were also most likely to have had histories of "hard" drug use (including heavy multiple-drug use and heroin addiction) and to have had early onsets of several types of drug use and criminal activity. Second, and perhaps more important, the information on an offender's drug history was more likely than official arrest records to be related to the amount and seriousness of self-reported criminal activity. As in the results of drug-crime studies discussed earlier, official arrest data were poor indicators of the type, amount, and severity of crime committed by these respondents.

These findings suggest a potential for using an individual's history of illicit drug use, including age of onset, in identifying high-rate, dangerous offenders. However, this approach has several limitations. First, a general caution is in order whenever findings based on aggregate data are applied to the individual case. Second, although self-reports of drug use and crime are generally valid when obtained from individuals who are either at large in the community, entering a drug-abuse treatment program, or already incarcerated, they tend to be less accurate for individuals being evaluated for initial disposition in the criminal-justice system. Approximately one out of every two new arrestees identified as drug users by urine testing conceal their recent drug use, even in a voluntary, confidential interview having no bearing on their correctional status.

(See also: Antisocial Personality ; Conduct Disorder ; Crime and Alcohol ; Families and Drug Use ; Family Violence and Substance Abuse )


Anglin, M.D., & Hser, Y. (1987). Addicted women and crime. Criminology, 25, 359-397.

Anglin, M.D., & Speckart, G. (1988). Narcotics use and crime: A multisample, multimethod analysis. Criminology, 26, 197-233.

Ball, J. C., et al. (1983). The day-to-day criminality of heroin addicts in Baltimore: A study in the continuity of offense rates. Drug and Alcohol Dependence, 12, 119-142.

Chaiken, J. M., & Chaiken, M. R. (1990). Drugs and predatory crime. In M. Tonry & J. Q. Wilson (Eds.), Drugs and crime. Crime and Justice: A Review of Research, Vol. 13. Chicago: University of Chicago Press.

Chaiken, J. M., & Chaiken, M. R. (1982). Varieties of criminal behavior. Santa Monica, CA: Rand.

Elliott, D. S., et al. (1989). Multiple problem youth: Delinquency, substance use, and mental health problems. New York: Springer-Verlag.

Gandossy, R. P., et al. (1980). Drugs and crime: A survey and analysis of the literature. National Institute of Justice. Washington, DC: U.S. Department of Justice.

Goldstein, P. J. (1989a). Drugs and violent crime. In N. A. Weiner & M. E. Wolfgang (Eds.), Pathways to criminal violence. Newbury Park, CA: Sage.

Goldstein, P. J., et al. (1989b). Crack and homocide in New York City, 1988: A conceptually-based event analysis. Contempory Drug Problems, 16, 651-687.

Goldstein, P. J. (1986). Homicide related to drug traffic. Bulletin of the New York Academy of Medicine, 62, 509-516.

Greenberg, S. W. (1976). The relationship between crime and amphetamine abuse: An empirical review of the literature. Contemporary Drug Problems, 5, 101-103.

Inciardi, J. A., et al. (1993a). Street kids, street drugs, street crime: An examination of drug use and serious delinquency in Miami. Belmont, CA: Wadsworth.

Inciardi, J. A., et al. (1993b). Women and crack-cocaine. New Yokr: Macmillan.

Inciardi, J. A. (1986). The war on drugs: Heroin, cocaine, and public policy. Palo Alto, CA: Mayfield.

Inciardi, J. A. (Ed.). (1981). The drugs-crime connection. Sage Annual Reviews of Drug and Alcohol Abuse, Vol. 5. Beverly Hills, CA: Sage.

Johnson, B. D., et al. (1991). Concentration of delinquent offending: Serious drug involvement and high delinquency rates. Journal of Drug Issues, 21, 205-229.

Kinlock, T. W. (1991). Does phencyclidine (PCP) use increase violent crime? Journal of Drug Issues, 21, 795-816.

Nurco, D. N., et. al. (1991a). A classification of narcotic addicts based on type, amount, and severity of crime. Journal of Drug Issues, 21, 429-448.

Nurco, D. N., et al. (1991b). Recent research on the relationship between illicit drug use and crime. Behavioral Sciences and the Law, 9, 221-242.

Nurco, D. N., et al. (1988a). Differential criminal patterns of narcotic addicts over an addiction career. Criminology, 26, 407-423.

Nurco, D. N., et al. (1988b). Nonnarcotic drug use over an addiction career: A study of heroin addicts in Baltimore and New York City. Comprehensive Psychiatry, 26, 450-459.

Nurco, D. N., et al. (1986). A comparison by ethnic group and city of the criminal activities of narcotic addicts. Journal of Nervous and Mental Disease, 12, 297-307.

Shaffer, J. W., et al. (1985). The frequency of non-narcotic drug use and its relationship to criminal activity among narcotic addicts. Comprehensive Psychiatry, 26, 558-566.

Tardiff, K., et al. (1986). A study of homicide in Manhattan, 1981. American Journal of Public Health, 76, 139-143.

White House Office of National Drug Control Pol-Icy. (2000). National Drug Control Strategy: 2000 Annual Report. Washington, D.C.

White House Office of National Drug Control Pol-Icy. (March, 2000). Drug Related-Crime Fact Sheet. Washington, D.C.

National Institute of Justice. (1999). 1999 Annual Report on Drug Use Among Adult and Juvenile Arrestees. Washington, D.C.

David N. Nurco

Timothy W. Kinlock

Thomas E. Hanlon

Revised by Frederick K. Grittner

About this article

Crime and Drugs

Updated About content Print Article