National Academies Press: OpenBook

Criminal Careers and "Career Criminals,": Volume II (1986)

Chapter: 2. The Impact of Substance Abuse on Criminal Careers

« Previous: 1. Issues in the Measurement of Criminal Careers
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 52
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 53
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 54
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 55
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 56
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 57
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 58
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 59
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 60
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 61
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 62
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 63
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 64
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 65
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 66
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 67
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 68
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 69
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 70
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 71
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 72
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 73
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 74
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 75
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 76
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 77
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 78
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 79
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 80
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 81
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 82
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 83
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 84
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 85
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 86
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 87
Suggested Citation:"2. The Impact of Substance Abuse on Criminal Careers." National Research Council. 1986. Criminal Careers and "Career Criminals,": Volume II. Washington, DC: The National Academies Press. doi: 10.17226/928.
×
Page 88

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

The Impact of Substance Abuse on Criminal Careers Eric D. Wish and Bruce D. Johnson OVERVIEW Approach This paper reviews what is known about how illicit drug use affects the pa- rameters of criminal careers, especially crime rates, and suggests directions that future research should take to fill the gaps in current knowledge about drug use and crime. To accomplish these goals, we have focused on the small number of studies of drug use that permit the com- putation of crime rates and that provide important implications for research. We have drawn heavily on our own research and that of our colleagues. In taking this approach, we have ex- cluded many excellent studies. The inter Eric D. Wish and Bruce D. Johnson are research staff members at Narcotic and Drug Research, Inc., New York City. Points of view or opinions in this paper do not necessarily represent the official posi- tion or policies of Narcotic and Drug Research, Inc. The authors would like to thank Lee Robins and Mary Toborg for their comments on earlier drafts of this paper. 52 ested reader may wish to refer to a num- ber of comprehensive reviews of the drugs and crime literature (Tinklenberg, 1973; McGIothlin, 1979; Weissman, 1979; Gandossy et al., 1980~. The reader should also be aware that our selection of studies influences the scope of our discussion and the applicability of our conclusions. Most of the studies we discuss concern crime among users of heroin and/or co- caine. These two drugs, along with aTco- ho! (which is reviewed separately in this volume), are the drugs that have been most frequently studied in relationship to crime. Although we briefly discuss mari- juana and phencyclidine (PCP), relatively few careful studies have been made of the relationship of these drugs to criminal behavior. We also discuss the relation- ship of barbiturate use and amphetamine use to crime, mainly in the context of studies that have focused on heroin or cocaine use. In focusing our discussion on studies of heroin and cocaine users, we have thereby limited the types of crimes and the types of populations that we report on. Also, because the use of heroin and

THE IMPACT OF SUBSTANCE ABUSE ON CRIMINAL CAREERS cocaine is rare in the general population and in persons uncler age 17, most of the detailed information about the relation- ship of these drugs to crime comes from studies of acIfflts who have been proc- essed by the criminal justice system or who have entered publicly funded drug abuse treatment programs. Finally, be- cause men are more likely to be arrested than women, most of the finclings refer to them. This is unfortunate in light of grow- ing evidence (Wish, Brady, and Cuadrado, 1984) that drug abuse may be more prevalent and severe among female arrestees than mate arrestees. Much of what we say is largely appli- cable to the indigent, less eclucatect, adult mate drug user who has been arrested. Our discussion is less relevant to the users of heroin and cocaine who are well educates! and legitimately employed (Washton and Gold, 1984; Zinberg, 19841; such persons are less likely to be found in state or federally funclec] treat- ment programs, from which many re- searchers select their samples. Little is known about the drug use and criminal behavior of these relatively affluent per- sons. However, a recent survey of 500 largely employed and educated persons who called a national hotline for help with cocaine-related problems (Washton and GoIc3, 1984) indicates crimes are less common among these persons than among less affluent users tv~icalIv stud- ied. Only 12 percent of the sample of mostly chronic cocaine users had been arrested for a cocaine-relatec3 crime, ancI 29 percent inclicatecT stealing from family, friends, or employers to support their habits. The fact that existing research does not permit more extensive discus- sion of drug use ant] crime among affluent populations should not hincler us from achieving our main purposes, however, since we are particularly concerned here with how drug use affects the criminal careers of persons processed by the crim S3 inal justice system, who are preponcler- antly not affluent. Summary of Findings We summarize here conclusions baser! on the information presented in this pa- ner. First, studies of persons who have been arrested and processed by the crim- inal justice system, of unapprehended criminals on the streets, and of persons in drug treatment programs indicate that as levels of illicit drug use (especially of heroin and cocaine) increase so does criminal activity (both drug-clistribution offenses and noncirug-related serious of- fenses). Second, among youths in the general population, the small subset who use cocaine, heroin, or pills for nonmedi- cal reasons account for a disproportionate amount of all juvenile crime. Third, per- sons in the United States who use these drugs enough to have associated legal problems tend to be so enmeshed in other deviance and adjustment problems as to make attempts to untangle the exact sequence of the onset of drug use and criminal behavior a futile and, perhaps, trivial pursuit. Fourth, chronic users of heroin and/or cocaine who are repeatedly arrested ant! processes! by the criminal justice system typically engage in a vari- ety of clrug-clistribution activities and other crimes. Fifth, treatment programs can reduce drug abuse and crime if the person remains in treatment. Ant! sixth, urinalysis appears to be an effective toot for identifying drug-using arrestees, but more needs to be learned about how to use this information. We have also attempter] to review a number of topics for which insufficient information was available to draw clefini- tive conclusions. Little is known, for ex- ample, about the natural course of drug use and crime among persons processed by the criminal justice system. Does in- carceration reduce or only delay drug use

54 and crime? Persons tend to relapse into c3 rug use and crime after release from treatment or detention, but c30 they make up for lost time? Even persons dependent on heroin have periods in their lives when they reduce or abstain from the use of drugs. More needs to be Earned about what brings these periods on and how they might be prolongecl. Drug use appears less prevalent among arrestees over age 35. Is this because ctrug-abusing criminals drop out of the active criminal population because of early death, incarceration, or institution- alization, or c30 they turn to alcohol or mature out of their drug use and criminal activities? Or do police avoid arresting oIcler criminals? Or is it that the older criminals, like the rest of the oIcler popu- lation, lack opportunities to use illicit clrugs? And, do these relationships apply equally to mate and female offenders? Much money and resources are being expenclect to reduce the supply of illicit drugs in the United States by seizing supplies and asking other governments to reduce poppy and coca plant production in their countries. These efforts assume that by reducing the supply one can re- duce the abuse of these drugs and the associated crime. Almost nothing is known, however, about how these efforts actually affect the crime rates of drug abusers. Do the higher prices for illicit drugs that result from a decrease in sup- ply lead to less use and therefore less crime, or floes the user merely increase his or her criminal activities to pay the increased prices, or is there no effect because the user turns to more abundant ([rugs? More needs to be learned also about how to reduce demanc! for drugs in of- fencler populations. Which offenders are the best cancticiates for intervention? Should major efforts go toward deterring the young, drug-using offender at risk of progressing to more serious drug abuse, CRIMINAL CAREERS AND CAREER CRIMINALS or toward deterring older persons, who may be more ready to change their ways? There is some evidence that court- ordered treatment may keep persons in treatment longer and, therefore, away from drugs and crime longer. More needs to be learned about how specific types of court-ordered interventions can reduce drug abuse and crime. The remainder of this paper expands on the points presented above. The paper is divided into two sections and two ap- pendices. In the first section, which is divided into 11 themes, we review the research and draw pertinent conclusions. In the second section, we present sug- gestions for future research on drug abuse and crime. Appendix A provides a sum- mary of many of the methodologic prob- lems involved in the study of drug use and crime that guided our review of the research. Appendix B provides criti- cal reviews of seven studies from which we have derived many of our conclu- s~ons. CONCLUSIONS FROM THE EMPIRICAL RESEARCH ON DRUG USE AND CRIME We have reviewed the studies indi- cated in Appendix B as well as other research bearing on drug abuse and crime and developed a set of themes to catego- rize the current state of knowledge. Each of these themes is discussed below. Drug Use and Crime Rates Among Youths and Adults After reviewing studies of individual crime rates conducted in the mid-1970s, Cohen (1978:229) concluded that, "clearly the most pressing research re- quirement for estimating the incapacita- tive eject is to provide adequate esti- mates of the individual crime rate (A)." These estimates, she added, should ac

THE IMPACT OF SUBSTANCE ABUSE ON CRIMINAL CAREERS count for variations in A by crime type, across the criminal population, and (lur- ing an individual career. This section pro- vicles information that shows how an of- fencler's use of hard drugs influences his or her crime rate. Two primary points are stressed: 1. Among youthful and aclult offenc3- ers, those who use hard drugs, especially if they are daily users, have higher crime rates than those who do not. 2. Studies of these active drug-using offenders that have measured self-report- ec! criminal behavior have produced esti- mates of crime rates that far exceed esti- mates based on arrest or conviction records. Stucties of Youths Information on the crime rates of cirug- using youths comes primarily from stud- ies basecI on data from Elliott and Huizinga's (1984) National Youth Survey (NYS). In assessing the value of such studies one must remember that serious criminal offenses are rare in the general population of youths and that analyses of the most deviant youths are necessarily based on a small number of subjects. Nevertheless, analyses of different types of offenders (e.g., those who limit them- seIves to minor offenses versus those who commit serious crimes) consistently show that use of serious "hard" drugs (primar- iTy cocaine or heroin) is associates] with higher rates of offending. Johnson, Wish, and Huizinga (1983) used NYS data to assess how rates of juvenile crime change according to the level of drug use and offender type (see also Elliott and Huizinga, 19841. Johnson and colleagues grouped youths into five classes of c3 rug use arranged hierarchi- cally (virtually all users of more serious drugs had used the less serious drugs) in terms of the seriousness of drugs used nonexperimentally in the previous year: 55 (1) no drug or alcohol use (N alcohol only used alcohol on four or more occasions (N = 5581; (3) mari- juana usecl on four or more occasions (N = 3011; (4) pills used on three or more occasions (N = 991; (5) cocaine-used on three or more occasions (N = 71, 12 of whom were heroin users). Mean annual crime rates were then caTculatecl for index offenses (rape, robbery, aggravated as- sault, burglary, larceny, and motor vehi- cle theft; homicide was excluded), minor offenses (thefts, assaults, vandalism), and drug sales. These finclings appear in Ta- ble 1. It is clear from Table 1 that the level of juvenile crime closely parallels the level of (lrug use. Both nonusers of drugs and alcohol and users of alcohol reported an average of only two or three crimes, most of them minor offenses, in the previous year. Youths who used marijuana had overall rates of crime that were three times higher than the rates for non-drug users or alcohol users. Youths who used pills but not cocaine, in turn, had higher crime rates than the users of marijuana or alcohol, particularly for index offenses and drug sales. The highest crime rates were found for the youths who reported the use of co- caine. Their rates of index and minor offenses were two to three times those of the pits users, and they had a very high annual rate (48) of drug sales. Separating the youths into offender groups based on the seriousness and number of crimes committed showed that even within these relatively homogeneous groups, youths who used pills or cocaine had the highest crime rates. In fact, one-fourth of the cocaine users had committed three or more index offenses in the previous year. Youths who used cocaine and committed multiple index offenses constituted only 1.3 percent of all youths but accounted for 40 percent of the index crimes reported by the entire sample. = 510~; (2)

56 CRIMINAL CAREERS AND CAREER CRIMINALS TABLE 1 Mean Annual Rates of Inclex Offenses, Minor Crimes, and Drug Sales in a National Sample of Youths by Level of Drug Use in the Prior Year (number of cases) Mean Annual Rate Index offenses Minor offenses Drug sales All offenses Youths Who in the Prior Year Used No Drugs/Alcohol Alcohol Marijuana (510) (558) (301) b 3 b 3 _b 2 b 6 1 8 Pills Cocainea/Heroin (99) (71) 3 12 9 24 9 21 48 78 Total (1,539) 9 NOTE: Some minor offenses (e.g., running away from home and skipping classes) have been excluded. aIncludes 12 who reported using heroin. bless than one crime per year. SOURCE: Johnson, Wish, and Huizinga (1983~. Conclusion. These findings from a study of a national sample of youths offer strong support for the hypothesis that se- rious drug use (especially of cocaine) and criminal offenses tend to be found among the same youths. These finclings are also consistent with other widely accepted studies showing that illicit drug use by youths tends to be accompanied by a variety of deviant attitudes and behaviors (lessor ant! lessor, 1977; Robins and Wish, 1977; Kandel, 19781. Studies of Adults A study of incarcerated persons in three states (Chaiken and Chaiken, 1982) fount! that violent predators, i.e., persons who reported committing robbery, assault, and drug dealing and who had very high crime rates, had extensive histories of drug use. Violent predators were more likely than others in the sample to have used hard drugs (including heroin) fre- quently as a juvenile and to have used drugs daily and in large amounts cluring the period studied (up to 2 years prior to the current incarceration). It is not clear from the data presented, however, whether cirug use is a major factor in differentiating crime rates among of- fender groups. "Robber-clealers," who committed robbery and drug dealing but not assaults, had Tower crime rates than the violent predators but similar drug-use histories (so far as one can tell, given that only significance levels are reported and not actual percentages) (Chaiken and Chaiken, 1982:Table 3.11. The robber- dealers had higher rates of participation for 15 of 19 juvenile and aclult drug-use measures. Compared with other inmates, both violent predators and robber-cleaTers had higher rates of juvenile heroin acIdic- tion, use of other hare! drugs as a juvenile, claily heroin use costing more than $50, daily barbiturate and amphetamine use (10 or more pills), ant! combined alcohol and amphetamine use. In the absence of more-cletaile(1 information, we have to conclude that the two groups, on the whole, hac! similar drug abuse histories and that the commission of assaults, not use of drugs, (differentiates the two groups. Furler information on this issue is provider] by a study (Chaiken, 1983) in which crime rates were computer! for these same sample members according to their offender group and level of illicit pill or heroin use during the study period. Table 2 presents the minimum estimates of crime rates for selected nondrug crimes, computed by truncating each per- son's annual rate for any offense type at

THE IMPACT OF SUBSTANCE ABUSE ON CRIMINAL CAREERS 365 (one per day). The findings of this study indicate that for each offender group high-cost heroin users had the highest crime rates. There was no mono- tonic relationship between drug use and rates of those crimes for other levels of drug use, but this may be because two important drug types (cocaine and mari- juana) were not measured. Even the vio- lent predators who did not report drug use had relatively high (156) crime rates, however. The findings indicate that ha- bitual use of heroin does tend to be ac- companied by high rates of nondrug crime, regardless of one's overall level or type of offending. On the other hand, the fact that violent predators who did not use drugs had high crime rates, even com- pared with some groups who were heavy daily heroin users, shows that serious heroin involvement is neither a necessary nor sufficient condition for high crime rates for nondrug crimes. (It should be noted, however, that the data presented in Table 2 omit the high rates of drug- dealing offenses among drug users. An- nual rates of drug dealing were generally greater than 1,000 among the high-cost heroin users, and in the 200 to 800 range for the other groups). Additional evidence for the link be 57 tween hard-drug use and crime rates is found in a recent study of 201 opiate users in Harlem (Johnson et al., 1985~. This study found that crime rates increased with the frequency of self-reported her- oin use. Daily heroin users (persons who used heroin 6 or 7 days a week) averaged 1,400 crimes per year; persons who used heroin less than 3 days a week averaged about 500 crimes per year. Although this finding could be affected by respondent- measurement problems, the subjects were interviewed daily for 5 days and then weekly so the recall period was short. If robbery, burglary, shoplifting, and other larcenies are taken as the index crimes, daily heroin users in this study committed 137 such crimes per year, and less regular heroin users committed 47 per year. Other crimes, e.g., forgery, pros- tit~tion, pimping, con games, and miscel- laneous nondrug crimes, were not related to level of heroin use among this group of users. The annual rates for drug-distri- bution crimes were much higher than for nondrug crimes, ranging from 245 (for irregular heroin users) to almost 900 (among daily heroin users). This study (Johnson et al., 1985) also used the offender typology developed by Chaiken and Chaiken (1982) and found TABLE 2 Annual Crime Rates for Robbery, Assault, Burglary, Theft, and Forgery- Fraud by Drug Use During the Measurement Perioc! Drugs Used No Pills Pills butLow-Cost Heroin Use Variety of Offender or Heroina No HeroinHeroin Use Over $50/Day Violent predator (N) 156 (50) 254 (76)134 (62) 326 (88) Robber-dealer 33 (32) 112 (51)156 (38) 194 (66) Low-level robber 27 (158) 19 (24)24 (26) 110 (23) Burglar-dealer 63 (62) 76 (50)127 (28) 184 (38) Low-level burglar 17 (89) 11 (23)5 (14) 78 (11) Property-drug offender 67 (52) 6 (21)104 (31) 204 (29) NOTE: Measurement period was up to 2 years prior to current incarceration. aStudy did not ask about cocaine or marijuana use during this period. Some of these persons could have used these drugs. SOURCE: Chaiken (1983).

58 that crime rates were generally high, con- trolling for offender type, for daily heroin users. These analyses, one should note, clid not control for age or other factors that could have affected the crime rates in each group. Table 3 presents the mean annual offense rates, their standard devi- ations, and the skewness for four offenses, for the 22 robber-dealers in the sample. (Robber-dealers were persons who com- mitted both robbery and drug dealing on 2 percent or more of their days on the street.) The offense-specific crime rates in Table 3 vary considerably. Burglary is a good example because the group was not defined on the basis of burglary rates. Although these 22 robber-dealers, as a whole, had an annual crime rate for bur- glary of 35.8, 6 of them committee! no burglaries, and another 4 committed more than 75. Thus, in computing and analyzing annual crime rates, one must pay special attention to the large variabil- ity that can be fount! even in a somewhat homogeneous group of drug-using of- fenders. Some of the variability in incliviclual offense rates may be explainer! by the finding that persons may have alternating periods of heavy and lesser drug use. One study reported high criminality during runs of narcotics use; when narcotics use declined so did crime rates (McGIothlin, Anglin, and Wilson, 1977~. Another study reported that addicts were six times more criminally active during periods of heavy narcotics use than cluring periods of CRIMINAL CAREERS AND CAREER CRIMINALS lesser use (Ball et al., 19811. Both of these studies offer further evidence of a link between heavy narcotics use and higher crime rates. A study of acldicts known to the police in Baltimore (Ball et al., 1981; Ball, Shaf- fer, and Nurco, 1983) has received consid- erable public attention because of find- in~s that show the magnitude of the increases in crime rates on days that per- sons used narcotics heavily, compared win days of less frequent use. The fincl- ings ofthis study are consistent with stud- ies reviewed above that document an increase in crime win increased narcot- ics use. However, because of problems of ambiguity in Me interview questions Mat measured the frequency of criminal activ- ity (noted in Appendix B), the exact esti- mates ofthe increase in criminal behavior should not be used as the basis for policy decisions until this study has been repli- cated in other sites. Another study (Wish, Klumpp, et al., 1980; Wish, 1982) analyzecl a 6-year re- cidivism file for 7,087 persons arrested in Me District of Columbia. Arrestees de- tectecl by urinalysis to be drug users (pri- marily morphine or phenme~azine) at any arrest cluring the period had an aver- age of 4.9 arrests during the 6 years, compared win an average of 2.7 arrests for persons not ~letected to be drug users. This result coup! have been observed because persons with multiple arrests probably had more urine tests during the period ant! thus a greater opportunity to TABLE 3 Mean Annual Crime Rates Among 22 Robber-Dealers Mean Annual Standard Offense Crime Rate Deviation Skewness Range Robbery 31.3 34.9 2.6 8-155 Burglary 35.8 50.4 2.3 0-212 Shoplifting 48.1 51.1 .9 0-144 Other larcenies 30.2 35.3 1.2 0-122 SOURCE: Johnson et al. (1985).

THE IMPACT OF SUBSTANCE ABUSE ON CRIMINAL CAREERS have a positive specimen and to be cIas- sified as a drug user. The researchers continuer] to fins! an association between the number of arrests and being a chug user, however, when they controller] for the number of urine test results available for each person. In a related analysis, persons found at an initial arrest to have a positive urine test had rates of multiple rearrest in a 4-year follow-up period that were significantly higher than those found for nonusers, after controlling for age and prior record (Forst and Wish, 19831. Similar findings were also reported in a study of male arrestees in Manhattan (Wish, Bracly, and Cuadrarlo, 19851. Ar- restees with a urine specimen that was positive in tests for any of four drugs (opiates, cocaine, PCP, or methadone; N = 2,647) at the index arrest had an aver- age of 3.5 arrests in a 3-year period, most of which occurred after the index arrest, compared with an average of 1.9 arrests for persons with drug-negative urine at the index arrest (N = 2,089~. This higher number of arrests was found among drug- positive arrestees of all ages. In addition, the number of arrests was related to the number of drugs found in a specimen. Arrestees with two or more drugs in their urine (N = 1,081) had an average of 4.6 arrests, compared with an average of 2.8 arrests Or persons with one drug (N = 1,5661. Thus, arrestees who had recently used multiple hard drugs (usually co- caine and heroin) had the highest number of arrests. Other studies of drug users (doss, 1976; Inciardi, 1979, 1984; Clayton anclVoss,1981; Collins ancIAllison, 1983; Johnson et al., 1985) and of the associa- tion between the price of heroin ant] levels of property crimes in the commu- nit,v (Silverman and Spruill, 1977) pro- vide evidence for a link between heroin and cocaine use and criminal activity. Conclusion. Studies that vary dramat- ically in the locales and populations sam 59 plecI, in the measures of crime and drug use, and in the cutting points and cIassi- fications of offenders and drug users have consistently found a strong association between the level of cocaine or heroin use and criminal behavior. Among the general population of youths and among adult offenders, users of these drugs have high rates of c3rug-distribution crimes and serious nondrug crimes, especially those that generate income. Daily users ofthese drugs tenet to have the highest crime rates. Demonstrating a link between serious drug use and crime is much easier than estimating the actual amount of crime committee] by drug abusers, however. Large estimates of the amount of crime attributable to heroin users have been challengecl by some as impossible and "mythical" (Singer, 1971; Renter, 19841. Diversity of Crimes Among Drug Users As in(licatecT above, recent research has demonstrated that some offenders who use hard drugs, like the violent predators, may have rates of violent crimes against persons that equal or exceed those found among offenders not using drugs. The analysis of the rates of arrest over a 6-year period for a sample of 7,087 arrestees (Wish, 1982), noted above, found that per- sons with a positive urinalysis test (at the time of at least one of their arrests) had rates of arrest for bait violations, larceny, robbery, burglary, and drug offenses that were two to three times higher than the rates for persons not detected to be using hard drugs. Drug users' rates of arrest for all other crimes were similar to those found for the nonusers. Analyses of consecutive arrests among drug users and nonusers from the same study showed a tendency for drug users to be rearrested for property crimes. A sample of all persons who had an arrest in

60 an 8-month panel period were selected for this analysis and all of their rearrests in the following 4 years were tracked. Each of 2,442 arrestees was classified as being drug positive (D + ~ by urinalysis at the initial arrest or drug negative (Do. The index arrest and the next arrest were classified according to six types of of- fenses: violent, robbery, property, victim- less, drug, and other. The results showed that for D- arrestees the next arrest was most likely to be for the same type of crime as the index arrest. Drug-negative arrestees initially charger] with a drug offense were an exception and were more likely to be rearrested for a property crime. All D + arrestees, however, were more likely to be charged with a property crime at rearrest than any other crime type, regardless of what the charge was at the initial arrest (Wish, Klumpp, et al., 1980:VII-22~. Ethnographic research of indigent drug users in New York shows that the ordi- nary, high-rate offender may switch from one crime type to another from one clay to the next ant! even on the same clay. For example, a person may commit a theft one clay, a burglary the next clay, several drug sales the next clay, ant! no crimes the next day (Johnson et al., 19851. Other studies of active street hustlers in Harlem have also suggested such a diverse pattern of offending (Strug, Stevie, et al., 1984; Strug, Wish, et al., 1984~. Although ethnographic studies of nonrandom samples of offenders provide findings with unknown representative- ness of other offenders, such studies do yield valuable insights into the link be- tween drug use and crime. For example, one of the reasons behind the variety and number of crimes that drug users report may be the rather modest amounts of money they earn from their crimes. Johnson et al. (1985) report that the aver- age nondrug crime committed by the re- spondents they studied netted the of CRIMINAL CAREERS AND CAREER CRIMINALS fender only $35 in cash; even the most lucrative nondrug crimes (burglary and robbery) netted an average of only about $80. Estimates of the annual criminal in- come from both drug and nondrug crimes ranged from $6,000 to $18,000. Conclusion Offenders with expensive drug habits clearly commit high rates of income- generating crimes, such as larceny, bur- glary, and robbery, in addition to high rates of drug-clistribution crimes. Evi- dence from ethnographic studies of indi- gent street users in New York indicates that these persons earn small amounts of money anti, thus, commit numerous crimes to finance their drug use. Drug Use and Violent Crimes If one considers robbery to be a violent crime, there is little doubt that drug users commit many violent crimes. However, there has been some controversy in the literature regarding whether drug users commit crimes specifically designed to harm persons (Wish, 1982~. Studies of the arrest charges for heroin-using versus nonusing arrestees have uniformly found that the heroin-using arrestees had higher proportions of arrests for property crime and Tower proportions of arrests for vio- lent crimes against persons (Kozel and DuPont, 19771. Similarly, inmates with a history of narcotic addiction were only one-thircl as likely to be serving a sen- tence for violent crime as were nonusers (Barton, 1976~. In reviewing this topic, McGIothTin (1979:361) cautioned against jumping to the conclusion that such results necessar- iTy mean that heroin users do not commit violent crimes: These findings have been loosely interpreted to conclude that narcotic addicts are less likely to commit crimes against persons than are

THE IMPACT OF SUBSTANCE ABUSE ON CRIMINAL CAREERS nonaddict criminals. Actually, the data do not warrant conclusions about the absolute fre- quency of crimes by He two groups. Addicts exhibit an especially high recidivism rate, and Be possibility that they commit many more property crimes, and some more violent crimes, than nonaddict criminals is not incon- sistent win the above results. In recent years the wisdom of this ob- servation has become clear. Analyses of a recidivism file for 7,087 arresters in the District of Columbia indicated that the percentage of arrest charges for violent crimes against persons for drug users (positive urine test) was lower than that for nonusers (Wish, 19821. However, the rates of arrest for violent crimes for drug- using ant] nonusing arrestees were equiv- aTent for assaults, sexual assault, and ho- micide. Rates of arrest among drug users for weapons offenses were higher than those for nonusers. And a study of incar- ceratec! persons (Chaiken and Chaiken, 1982; Chaiken, 1983) found that many of the violent preclators, the group cleaned by drug cleating, assault, and robbery, were heroin users, many of whom had expensive habits. Drug-using offenders, especially those involved in drug-clistribution activities, may be especially prone to commit crimes against persons. Several jurisclic- tions have recently issued statistics that indicate that between 20 and 30 percent of their homicide cases appear to be "drug related," that is, the victims or perpetrators were either drug users or dealers (Goldstein, 1985; Heffernan, Mar- tin, and Romano, 1982; McBride, 19831. And toxicologic studies of homicide vic- tims in New York have shown a high prevalence of alcohol and drug use by the clecedents (Haberman and Baclen, 19781. A rationale for the prevalence of vio- lence among drug abusers has been sug- gestecT in terms of a "systemic moclel" (Goldstein, 1985~. This model holds that the drug-clistribution system relies main 61 Ty on violence and its threat to maintain "orcler" and to control the sale of these valued, but illegal, substances. A variety of expectations of violence have been developer! by higher level dealers to keep Tower level distributors "in line." And lower level users-dealers see drug distributors as prime candidates to "rip off,' (rob or burglarize). Distributors who have been victimized rarely report such crimes to the police; they settle the mat- ter themselves. Conclusion Users of heroin have, in the past, been considered to be unlikely to commit vio- lent crimes against persons. Recent stud- ies suggest that hard-drug users commit violent crimes at least as often as nonus- ing offenders. The pervasive violence in the drug-distribution system may even increase the likelihood of drug users' be- coming perpetrators or victims of violent crimes. More research is needed to clarify the hypothesized link between violent crimes and drug-distribution activities. Drug-Distr~bution Activities ant! the Measurement of Crime Rates There are a number of reasons for sepa- rating drug-related crimes (e.g., possession or sale) from the computation of rates of crime. Drug users, by definition, commit- ted drug-related crimes. However, drug- distribution activities are so much a part of the daily lives of drug-involved of- fenders that to ignore these activities is to underestimate their crime rates seriously. Virtually all studies of high-risk popu- lations have found that the rates of drug selling exceed those of any other offense type, especially for users of cocaine or heroin. Chaiken and Chaiken (1982) found that their subjects reported be- tween 90 and 160 drug sales per year. Even among persons who were not daily

62 heroin users, the number of drug sales (about 100) exceeded the number of thefts by a factor of two to five. Drug users in Miami reported two to three drug sales for every theft that they committed (Inciardi, 19791. And others have fount! that approximately 10 percent of Ameri- can youths sell drugs in any given year, and a few do so more than 50 times per year (Single and Kanclel, 1978; Clayton and Voss, 1981; Johnston, Bachman, ant! O'Malley, 1982; Elliott et al., 1983~. Daily heroin users in East Harlem reported committing an average of 1,000 cirug- ctistribution crimes per person per year (Johnson et al., 1985~. Conclusion Drug-ctistribution activities must be taken into consideration when measuring the rates of crime among drug users. These crimes are among the most com- mon committed by drug users, and poli- cies of selective incapacitation or treat- ment of drug users may have their greatest impact on these crimes. Onset of Drug Use and Crime: Does It Matter Which Occurred First? The onset of drug use ant! crime has been given considerable attention in the research literature. There is often an im- plicit assumption that knowing when in the life cycle ant! in what order the two types of behavior first occur may help to resolve two issues: (1) how to intervene in and prevent these behaviors and (2) whether the onset of drug use changes a person's level (or type) of criminal behav- ior. Persons who begin to use drugs or alcohol at an early age have a greater likelihood of having problems with sub- stance abuse and alcoholism as adults. The evidence is less definitive on the issue of whether drug use precedes or follows onset of criminal behavior, and it CRIMINAL CAREERS AND CAREER CRIMINALS appears that this relationship may depend on the availability of the drug and the con- ventional age at which its use is initiated. The typical addict studied before 1950 lic! not have a prior criminal background (Greenberg anct Adler, 19741. These per- sons, predominantly rural, white south- erners, became actdicted in their middle twenties, usually as a result of using pre- scribed drugs. Around 1950 a shift oc- currec3 in the type of persons who became heroin aclclicts. Adclicts were now urban blacks and Spanish-speaking males who voluntarily used heroin and who had a history of criminality prior to the begin- ning of adcliction in their teenage years (DuPont artcl Kozel, 19761. The weight of the evidence seems to support the con- clusion that currently most (not all) users of heroin and other hard drugs who even- tually come to treatment programs or who are apprehencled by the police have cle- viant or criminal backgrounds that pre- cecled their addiction. Heavy use of her- oin and injection of heroin and cocaine tend to begin in the late teens or the early twenties (Inciardi, 1981; Clayton ant] Voss, 19811. Once aclclicted, these per- sons become more involves] in drug-dis- tribution activities and other income-gen- erating crimes (McGIothTin, 19791. Heroin use and, to a lesser degree, cocaine use have a bad reputation in American society, and there is consicler- able self-selection involved in the use of these drugs. Persons who are deviant in childhoocl are more likely to use these drugs, and consequently, it is cliff?icult to cietermine how many crimes committee! by users are the result of an underlying disposition toward deviance and criminal behavior. This is a major problem in as- sessing the causal role of drug use in crim- inal behavior. After considering these is- sues, Robins (1979:328) concluclecl, Thus, while it is true that the kinds of people who use heroin are also likely to commit crimes, and that committing crimes makes

THE IMPACT OF SUBSTANCE ABUSE ON CRIMINAL CAREERS them especially likely to come to public atten tion as addicts, Me fact Mat Me number of property crimes does seem to fluctuate with Me use of heroin makes it highly probable that addiction does directly increase the frequency of ~efc and other crimes designed to provide money for drugs. Since Robins's review, a number of stud ies (Ball et al., 1981; Goldstein, 1981, 1983; Wish, Numpp, et al., 1981; Chaiken ant] Chaiken, 1982; Ball, Shaffer, and Nurco, 1983; Chaiken, 1983; fohnson et al., 1985), many of which we have reviewed in this report, have shown the selling drugs in Harlem (Wish, Anderson, huge repertoire of deviant behaviors in et al., 1984) also found that younger ar which harcI-drug users are involvecI. restees are unlikely to have test results that are positive for opiates. Urine tests revealer] opiate use for 22 percent of the persons age cl 18 to 20 and for 33 percent of the persons aged 21 to 25, compared with rates above 55 percent for persons 26 or older. However, cocaine was cletected in arrestees of all ages. Injection of co caine was rare among young arrestees who reported using the drug. Only 15 percent of the cocaine users under 25 who had used cocaine in the 48 hours prior to arrest (N = 42) had injected the drug, compared with 60 percent of per sons aged 26 to 30 and 88 percent of those ages] 31 to 35. These age-relatecT trencis based on a sample of drug dealers arrested in Harlem were also observed in a study of serious offenders arrested throughout Manhattan (Wish, Bra(ly, et al., 1984~. The sample consisted of more than 6,000 men arrested for serious offenses (approx imately two-thir(ls were charged with fel ony offenses; 20 percent were charged with sale or possession of ([rugs) and processed in the Manhattan central book ing facility between March and October 1984. As one would expect, the preva lence of positive urine tests and self reported drug use was lower among this population of persons arrested for a vari ety of offenses than among the drug deal ers in Harlem. 63 of Columbia (Wish, Klumpp, et al., 1980) found that in each of the 5 years from 1973 through 1977 the likelihood of a positive urine test for hard drugs (usually morphine) was Tow for arrestees below age 20 and peaked for those in their thirties. However, recent research has in- clicated that the types of urine tests user! at that time in the District of Columbia were largely ineffective in detecting co- caine (Wish, Strug, et al., 1983), a (1rug commonly found among young arresters. A study of 110 persons arrested for Conclusion Untangling the causal nexus between drug use and crime is, perhaps, an impos- sible and unproductive enterprise. Re- searchers examining this question using information from the National Youth Sur- vey conclucled, "the concern over the temporal sequences of these two prob- lems in an effort to determine causal pri- ority is misdirected. If they have a com- mon etiology, either may precede the other with no causal implications" (El- liott and Huizinga, 1984:961. It appears that heavy use of hard drugs is an excel- lent indicator of persons who have high rates of criminal behavior. Whether they began using drugs before or after they committee] their first nonc3rug crime is probably a function of opportunity and other societal factors. Of more practical consequence is the question of how one might intervene in the process of devel- oping deviant behavior at an early age. Dug Use Among Arrestees Studies using urine tests to identify recent drug use provide some indication of the prevalence and development of drug use in offender populations. A sturly of 57,000 persons arrested in the District

64 Response rates were high in the study (Wish, Bra(ly, et al., 1984~. Ninety-five percent of the persons approached agreed to be interviewed about their drug use and treatment histories, ant] 80 percent of the respondents provided a urine speci- men for analysis. The percentage of ar- restees at each age level who had a urine test positive for any of four drugs (co- caine, opiates, methadone, and PCP) is shown below: Age of Arrestee (N) 16 (185) 17 (179) 18 (203) 19 (224) 20 (254) 21 (248) 22 (262) 23 (238) 24 (241) 25 (213) 26-30 (1,005) 31~5 (653) 36+ (915) Percent Drug-Positive 34 36 49 50 47 59 63 63 54 65 64 64 48 The likelihood of a positive urine test tended to increase with the age of the arrestee, through age 35. A large decline in the rate of positive tests occurred for offenders over age 35. The age ofthe arrestee was also related to the type of drug cletectecI. At all age levels, cocaine was the drug most likely to be detected. However, for persons be- Tow age 21, opiates and methadone were rarely cletected. Between the ages of 16 and 25, PCP was detectect at about a uniform rate (mainly between 16 and 25 percent); few arrestees above the age of 25 (12 percent or less) had tests positive for PCP, however. The probability of de- tecting multiple drugs in the urine in- creased with age. Between 7 and 10 per- cent ofthe arrestees uncler age 21 had two or more c3 rugs in their urine, compared with between 17 and 32 percent of oIcler arrestees. Arrestees who were between the ages of 25 and 35 hac! the most drugs in their urine. One-half of their speci CRIMINAL CAREERS AND CAREER CRIMINALS mens contained cocaine, 30 percent con- tained opiates, 10 percent methadone, and 10 percent PCP. The fact that arrestees uncler age 20 were unlikely to have tests positive for methadone and opiates and were less likely to have multiple drugs in their urine suggests that these persons hack less severe drug problems than oIcler ar- restees. Information from their interviews indicated that these younger arrestees were less likely to report current depen- dence on drugs or the need for treatment than were oilier arrestees. In addition, these you~fuT users of cocaine were more likely to report that they snorted the drug than older arrestees, who tenciec! to inject cocaine, often with heroin. The generaTizability of these two stud- ies of arrestees in New York City to ar- restees in over cities is unknown. Paral- leT findings are emerging, however, from a comparable ongoing study of urine test- ing of arrestees in Washington, D.C. (Toborg, 1984). In both sites approxi- mately 56 percent of all tested arrestees have a urine test positive for drugs. The prevalence of tests positive for opiates in the two cities is the same, approximately 20 percent. However, the prevalence of cocaine in Washington is about one-half the prevalence in New York (42 percent), and the prevalence of PCP is more than twice as great in Washington. Thus, the overall level of drug involvement among arrestees in the two cities is the same, although the preferred drugs appear to differ. Given the caveat regarding their gene- ralizability, these findings raise some in- teresting points. First, it appears that by the time persons are eligible for arrest as an adult (16 in some jurisdictions, 18 in others), there is almost a 50 percent chance that the person is already using an illicit drug, for example, cocaine or PCP. (These urine test results should probably be considered conservative estimates of

THE IMPACT OF SUBSTANCE ABUSE ON CRIMINAL CAREERS drug use, given that many drug-using arrestees probably did not use a drug close enough to the time of arrest to pro- vide a specimen capable of revealing it. The tests can generally detect opiate and cocaine use within 24 to 48 hours and PCP use within five days.) In addition, involvement with serious drugs may be a developmental process, beginning with snorting of drugs, followed by injection of one or more drugs. It is not possible to tell whether the differences in drug use and detection for arrestees at different ages are the result of differences in cohorts or of true developmental stages. The self- report information on the age of initia- tion of these behaviors among the entire sample can be used to examine this ques- tion. Finally, the findings indicate a de- crease in recent drug use by arrestees over the age of 35. Although the idea that persons may mature out of drug addiction has been suggested (Winick, 1962), it is also possible that this decrease in drug users in the population of arrestees could be the result of such things as drug users' avoidance of arrest, imprisonment or in- stitutionaTization, death, abstention from drugs, or alcoholism or some other ill- ness. We are aware of no studies of of- fender populations that indicate which of these factors are pertinent. Conclusion Current studies of drug use in two large eastern cities show that a little more than 50 percent of arrestee populations gave evidence of recent illicit drug use. The order of the prevalence of drugs in the two cities differed, however. There was some evidence from two studies that ar- restees under age 21 were less involved with opiates and cocaine use by injection than older arrestees. A decrease in the prevalence of recent drug use among ar- restees over age 35 raises important ques 65 lions regarding the ultimate course of serious drug use among offender popula- tions. Marijuana and Crime A Weak Link Research on the relationship between marijuana use and crime has generally found little evidence that the drug in- duces any type of criminal behavior other than, possibly, selling the drug. Youths have reported that marijuana use reduces their inclination toward violent behavior (Tinklenberg, Roth, et al., 1976; Tinklen- berg, Murphy, et al., 19811. And other research has shown that marijuana use takes place as part of an unfolding devel- opment of other problems and noncon- forming behaviors. In his review of the correlates of marijuana use lessor (1979: 348) endorses the view that "delinquency and marijuana use are manifestations of the same phenomena involvement in deviance or problem behavior and are associated with each other by virtue of a common relationship to social, psycho- Togical, and economic etiological vari- ables." One of the difficulties in assessing the role of marijuana use in crime is that use of other drugs often follows shortly (Kandel, 19841. Studies have shown that .. . `` ,, marijuana Is a gateway c .rug it opens one up to the use of other drugs Johnson, 1973; Robins and Wish, 1977; O'Donnell and Clayton, 19811. To be sure, only some marijuana users go on to use harder drugs, but the risk certainly increases with in- volvement with marijuana. A current study of clients classified by treatment programs as having marijuana as the pri- mary drug of abuse found that, with few exceptions, those persons were using a variety of other drugs, including PCP and cocaine (Kleinman et al., 19841. Because most daily marijuana users also use other drugs, heavy marijuana use and multiple drug use are confounded, and attempts to

66 isolate the impact next to impossible. Conclusion r .. or marijuana use are The role of claiTy marijuana use in seri- ous crime is badly confounded by the use of more serious drugs. Marijuana use may provide an introduction to the illicit drug market and to the use of drugs, such as heroin and cocaine, that have a more direct role in both c3rug-relatec3 and nondrug crime. PCP and Violent Crime A Stronger Link Attempts to measure the prevalence of the use and impact of PCP are hampered by the fact that the substance is often distributecI under a variety of names and misrepresenter! as other drugs. Because marijuana is sometimes laced with PCP, persons may be unaware that they have consumed the latter. Previously, PCP's chief use was in veterinary medicine as an anesthetizing agent, although it was originally synthesized for use with hu- mans (Peterson and StilIman, 19781. It can be taken orally, smoked, snorted, or injected. PhencycTidine is often classified as a hallucinogen, although, because of its diverse actions there is some disa~ree ment as to how to classify it. Small closes of PCP lead to a drunken state, and larger doses can produce anesthesia, convul- sions, and a psychotic-like state. The available literature on PCP use and crime is sparse and consists mainly of case studies of persons who have commit- tec3 violent, often bizarre acts (Siegel, 1978, 1980; Fauman and Fauman, 1982~. Death from cirowning, often in small amounts of water. has been a frequent cause of PCP-related death in California ant! the media's emphasis on such events helps to give PCP a bad reputation. CRIMINAL CAREERS AND CAREER CRIMINALS PhencycTidine is easily synthesized and inexpensive. One would therefore not expect to find the increases in in- come-generating crimes with the use of PCP that are found among users of expen- sive drugs. The potential link between PCP use and violent crimes against per- sons is based on the idea that some per- sons become so disoriented when using the drug that they commit acts for which they are not responsible. These assump- tions are reflected in the debate regarding the viability of the legal defense of dimin- ishec3 capacity for crimes committed clur- ing PCP intoxication (BaxTey, 19801. It has been arguer! that persons who have committed violent crimes while uncler the influence of PCP are not legally re- sponsible for their acts because they have an inability to have criminal intent. Sie- ge] (1978:285) has concluded, The PCP-intoxicated user's orientation toward the immediate present and disregard for long range consequences of his/her behavior would make it difficult for him/her to premeditate criminal acts. But the tendency to react strongly to sensory stimuli in the immediate environment, the inclination to refer every- thing to oneself that often develops into para- noia, and the need to do something due to intense psychomotor stimulation can all pro- duce an aggression-prone individual. Once again it must be emphasized that emotionally stable people under the influence of low doses of PCP probably will not act in a way very differently from their normal behavior. Epidemiologic evidence of the per- centage of all PCP users who become intoxicated with the drug and commit violent acts is not available. As the state- ment above implies, one would suspect that only a small minority of PCP users ever reach such a stage. In ongoing stud- ies PCP is the (lrug that is most likely to be cletecte(1 in the urine of arrestees in Washington, D.C.-found in 30 percent of mate and female arrestees (Toborg,

THE IMPACT OF SUBSTANCE ABUSE ON CRIMINAL CAREERS 1984), but it is less prevalent in the urine of arrestees in New York City 12 per- cent of male arrestees (Wish, 19861. If only 1 percent of these PCP-using ar- restees in Washington and New York ex- hibited the bizarre, violent behavior at- tributed to PCP-intoxicated persons, our institutions would be overwhelmed with them. One study of drug use anct violence among 112 boys committed to a training school (the average age was 15) clid find a strong relationship between PCP abuse and offenses against persons Kimonos and Kashani, 19801. A psychiatrist inter- viewed each youth and rated each one's abuse of 13 types of ([rugs. The number of prior offenses against persons (e.g., rape, assault, robbery) and of property offenses was obtained from juvenile records. Abusers of PCP had a greater number of offenses against persons (mean = 15.3 offenses) than abusers of any other drugs. Abusers of barbiturates hac3 the second highest number (mean = 10.0), followed by abusers of amphetamines (mean = 5.2~. The level of PCP use was not signif- icantly related to rates of property of- fenses. Because many of the boys abused multiple drugs and only 17 percent of the offenses against persons were preceded by use of a drug within the prior 24 hours, it was not possible to attribute the violent offenses to the use of any of the drugs. Conclusion PCP is used by persons who tend to be multiple drug users. It is one of the more common drugs used by arresters, al- though its prevalence varies considerably by jurisdiction. An unknown, but proba- bly very small percentage of users suffer extreme PCP-incluced intoxication and disorientation and commit bizarre, often violent acts. Much more research is needed to identify the extent of these 67 problems in users, and to learn how per- sonaTity, other drug use, and the quality and quantity of PCP ingested contribute to the occurrence of violent behavior. The Role of Hard-Drug Use in Crime Despite the strong link between drug use and crime clocumented in the previ- ous sections of this paper, there is a Rearm of literature examining the specific nature of that link. At best, some surveys have asked persons to recall whether they were under the influence of drugs or alcohol at the time they committed a crime. Examples of questions that have not been addressed are: Were you high or experiencing withdrawal symptoms dur- ing the crime? When and which sub- stances were taken, and with whom? How clip the drug affect the crime? What drugs were taken after the crime, and how soon? The absence of such informa- tion may be clue to the fact that obtaining such event-specific details requires that persons be interviewed as soon after com- mitting a crime as is possible. Much ofthe iscussion in this section is about recent studies that have used this approach. We Took, first, at when the drug use occurs, and then the role of the drugs in the crime. According to a survey sponsored by the Bureau of Justice Statistics (1983a,b) of 274,564 prisoners in state correctional fa- cilities in 1979, 33 percent reported being under the influence of an illicit drug (in- cludes 8 percent who used marijuana only) at the time they committed the crime for which they were incarcerate Other studies have asked youths about the commission of delinquent acts and the role of ([rugs that were used e.g., Elliott and Huizinga's (1984) National Youth Survey and Tinklenberg and col- leagues' study of youths in California (see Tinklenberganc3 Ochberg, 1981; Tinklen

68 berg, Murphy, et al., 1981). However, because use of hard drugs (cocaine and heroin) and commission of serious crimes are rare in persons under age 18, surveys of the general population can contribute little to an understanding of how often drug use is concurrent with the commis- sion of a crime. To gain information on this topic, one must look at samples of high-risk youths or persons who have had some contact with the criminal justice system. One such study found that juve- niles admitted to a training school re- ported taking drugs, primarily alcohol and marijuana, to give them courage for committing violent crimes (Simonds and Kashani, 1980~. Studies that have tested urine speci- mens from arrestees in New York City and in Washington, D.C., as noted above, have found that a little more than one-half of all arrestees had recently used an illicit drug (Toborg, 1984; Wish, Brady, et al., 19841. A study of 116 arrestees charged with the possession or sale of illicit drugs found even higher rates of recent drug use (Wish, Anderson, et al., 19841. Eighty percent of the persons interviewed within 1 to 4 hours of arrest had urine specimens that were positive for hard drugs, usually heroin or cocaine, which indicates that the drugs were probably used within the past 24 to 48 hours. There was no indication in either of these stud- ies whether the drug use played a role in the instant arrest. A number of studies of the role of drug and alcohol use in the crime event have been undertaken by staff at the Interdis- ciplinary Research Center (IRC) in New York City. These studies used ethno- graphic techniques to recruit persons in East Harlem who had recently commit- te(1 a nondrug crime. These studies of unapprehended persons can provide some insight into why ant! when active . , . CRIMINAL CAREERS AND CAREER CRIMINALS unknown probability of selection, the generaTizability of the findings is un- known. An IRC study of 59 unapprehencled street criminals found that alcohol was frequently taken in large quantities be- fore a crime, often to facilitate the crime by calming nervousness (Strug, Wish, et al., 19841. Cocaine and heroin were also taken for these reasons. These 59 persons reported committing 103 nondrug crimes in the 36 hours prior to their interviews. Theft, robbery, shoplifting, and burglary were the most common crimes. Alcohol was found to be the drug most likely to be used at the time of the crime (in 37 per- cent of the 103 crimes), and respondents reported that alcohol actually helped them to perform their crimes. Most of these persons had no cocaine or heroin to take before the crime and user! their crim- inal income to purchase drugs and aTco- ho! shortly after the crime. Conclusion The evidence is scanty regarding the exact timing of drug use en c] crime. Infor- mation from urine-testing programs tells us only that about one-half of arrestees in two eastern cities had used a drug some- time near the arrest. Whether other cities have this (legree of drug use among ar- restees is unknown. Findings from stucI- ies of active criminal drug users do in(li- cate that alcohol and other drugs may be used to prepare for a crime and are almost surely used after the crime, if money has been obtained. The generaTizability of these findings to other offenders is un- known, pencling replication ofthe studies in other sites. Interventions for Reducing Drug Use and Crime criminals cane chugs. However, since per-There is a considerable body of litera sons were selected for study with anlure on the efficacy of drug abuse treat

THE IMPACT OF SUBSTANCE ABUSE ON CRIMINAL CAREERS ment programs for reducing clients' crime rates (SelIs et al., 1976; Ganclossy et al., 1980; Collins, Hubbard, and Rachal, 1985; Collins, Rachal, et al., 1982a,b; Col- lins and Allison, 19831. To review this literature would be beyond the scope of this paper. However, whether treatment appears to reduce drug abuse anc3 crime has implications for policies that involve involuntary detention for treatment. By far, the best-designecl anc3 best-ex- ecuted study of a treatment population is the evaluation of the California Civil Ac3- ctict Program for persons convicted of cer- tain crimes or recommencled for treat- ment by the district attorney (McGIothlin, Anglin, anc3 Wilson, 1977; Anglin anc3 McGIothTin, 1984~. Until 1970, this pro- gram consisted of a 6-month inpatient period, followed by a 3-year, outpatient c3rug-treatment period. Inpatient treat- ments incluclec3 group therapy, a modi- fied therapeutic-community approach, anc3 educational anc3 vocational training. The outpatient treatments incluclecT su- pervision by a parole agent, weekly coun- seling, and drug surveillance through uri- nalysis or naTine, an antinarcotic that brings on withcirawal symptoms if the person has been using opiates. If a parol- ee's whereabouts were unknown for 72 hours, the parole officer issued a warrant for the person's arrest and return to inpa- tient treatment. In 1970 the inpatient pro- gram was eliminated, outpatient treat- ment was reducer! to 2 years, and a methadone outpatient component was in- stitutecI. Using self-report measures anc3 arrest records, it was found that, compared with a group of persons who were released from the program early because of a stat- utory change, persons who participated in the outpatient program that inclu(led strict supervision had a significant recluc- tion in narcotics use while they were in the program. In addition, both self- reports and arrest records confirmed a 69 drop in non(lrug criminal activity during this period. Interventions that included supervision with urine testing resulted] in Tower rates of daily narcotics use, drug dealing, and other criminal activity, an higher rates of employment than (lid su- pervision-without-testing or no-super- vision statuses. The study concluder] that the outpatient program with heavy super- vision of parolees was effective because it reclucecl, not eliminated, the person's daily runs of narcotics use. The report (McGIothlin, Anglin, and Wilson, 1977: 71) states that "a policy of containment aimed at limiting the extremes of narcotic usage and its associate(1 criminal behav- ior can be successful in minimizing the social costs of acldiction, although per- haps not achieving the traditional goal of abstinence." Because the use of methadone to treat addicts became popular in California dur- ing the study period, it was possible to analyze its impact on study members. While almost half the persons used nar- cotics daily when not receiving metha- clone, only 6 percent used narcotics daily when they were receiving methadone. A reduction in criminal behavior from 42 to 14 percent was also found for those re- ceiving methadone. Improvement in em- ployment was less marked, and alcohol use was slightly greater while on metha- clone. While McGIothTin and colleagues found that methaclone-treated clients die! as well as those who received the super- vision-with-testing treatment, another rigorous study reporte(1 less favorable re- sults. For adcLicts receiving methadone treatment, Lukoff (1974, 1976) reported a great reduction in drug crimes but not in noncirug crimes. Lukoff concluded that more time in treatment was neecled to bring about the profound changes in life- styTes and habits nee(le(1 to produce large reductions in criminal behavior. Another study (Hunt, Lipton, and Spunt, 1984)

70 found that methadone clients committed fewer serious crimes in treatment than clic] narcotics acIdicts not in treatment. It may be that differences in types of clients ant] programs were partially a factor in the different outcomes reported by these studies. As we note in the next section, more research is neecled to determine which interventions are effective with particular types of offenders. Using ciata from a large study (Treat- ment Outcomes Prospective Study) of persons entering feclerally funded treat- ment programs in 1979 and 1980, another study (Collins and Allison, 1983) also found that the use of legal pressure to induce entry into a treatment program may have beneficial impact on addicts. Collins and Allison studied arrestees who were divertecI through the Treatment Al- ternatives to Street Crime (TASC) pro- gram to outpatient drug-free programs and residential treatment programs. Their findings indicate a statistically significant increase in retention in treatment for per- sons who were referrer! to treatment by the criminal justice system as opposed to other means. Because legal pressure had a greater impact than other referral meth- ods on persons entering a residential treatment program anct because TASC referrals stayed longer than non-TASC criminal justice referrals in outpatient clrug-free programs, Collins and Allison (1983:1148) conclude that the findings are consistent with the interpretation that "legal pressure is most effective when accompanied by monitoring or surveil- lance of clients' behavior." Conclusion The studies reviewed indicate that of- fenders apprehen~lecl by the criminal jus- tice system may be helped if they are mandated to participate in treatment that is accompanied by strict supervision and drug surveillance over prolonged peri CRIMINAL CAREERS AND CAREER CRIMINALS oafs. More research is needed, however, to determine the types of interventions that work best with particular types of offenders ant! with persons at arrest, pro- bation, and parole stages. Iclenti*ing High-Rate Drug Users To intervene with cirug-using offend- ers, it is necessary to identify them. Al- though self-report information on prior drug use has been shown to be valid in research studies, it is unlikely that ar- restees would disclose such self-incrim- inating information if it were to be usecl in making decisions regarding their cases. Prior research (Wish, Klumpp, et al., 1980) has unclerscore(1 the fact that one cannot reaclily identify drug use by ar- restees through self-report in a cell-block interview or by type of arrest charge. In an analysis of 17,000 arrest cases from 1973 and 1974 Wish and colleagues founcI that only 10 percent of the arrestees who were positive for har(1 (1rugs by urinalysis were charge(1 with a (drug offense. Other analyses in(licatecT that 50 percent or fewer of the arrestees who were positive by urine test acimittecl in a cell-block interview that they used hard ([rugs. A stu(ly of arrestees in Manhattan (Wish, Brady, et al., 1984) has confirmed the finclings that compared with their uri- nalysis results, arrestees greatly under- report their use of PCP, cocaine, and opiates during the 24 to 48 hours prior to arrest. This underreporting occurred de- spite the fact that the arrestees were told that the research interview was voluntary and anonymous, that it wouIc3 not affect their cases, and that all information was protected from subpoena and use in civil anti criminal proceedings by a Federal Certificate of Conficlentiality. When com- parable offenders were interviewed in a storefront in East Harlem, the concor- dance between self-reports and urinalysis results ranged between 70 and 80 percent

THE IMPACT OF SUBSTANCE ABUSE ON CRIMINAL CAREERS (Wish, Strug, et al., 1983; Wish, Brady, and Cuadrado, 1985), and most discrepancies still resulted from underreporting of the drugs that were detected in the specimens. In another study (Wish, Anderson, et al., 1984) in New York City 116 persons arrested in Harlem, primarily for posses- sion or sale of hard drugs, were inter- viewed within 2 to 4 hours of arrest and a urine specimen was obtained. The inter- view information and the urinalysis re- sults indicated considerable drug use and other problems. Almost one-half (46 per- cent) claimed to have been physically de- pendent on heroin at one time in their lives. One-quarter reported that they were currently dependent on the drug, and one-third claimed that they needed drug abuse treatment at the time of arrest. The urine tests indicated that almost 80 percent of these persons were using drugs, mainly cocaine and heroin. Fifty-four of the persons whose cases had been disposed by the time of the analysis had pled guilty, and of the 30 sentenced to some jai] time, 76 percent were released immediately for time served or given a sentence of 29 days or less. On release, 43 of the respondents appeared at the research storefront in Harlem to be interviewed in greater de- tail. (Respondents were paid for their ini- tial interview and urine specimen if they contacted the research staff after release. At that time they were asked to partici- pate in a longer interview and to provide a second specimen.) Compared with the persons who were not reinterviewed, reinterviewees were older and more likely to report that they currently needed treatment for drug or alcohol use (64 ver- sus 27 percent). The educational and eth- nic backgrounds of the two groups were the same, as were their reports of daily drug use within the past month. The follow-up interviews and the urine specimens at the second interview indi- cated that many of the persons who had 71 been released had resumed to drug use and crime almost immediately. Approxi- mately 80 percent of the second speci- mens were positive for hard drugs, and the majority for more than one drug. These persons had been apprehended and released soon after arrest and there was no systematic attempt by criminal justice system staff to measure or inter- vene in their drug use. If all arrestees had the high rates of drug-positive urine tests (heroin or co- caine) that were found among these drug dealers from Harlem, a strategy of using urine tests to flag potentially high-rate offenders would be impossible. (A policy of selective incapacitation is based on the idea that only a small proportion of of- fenders exhibit very high rates of crime.) Current studies of large numbers of ar- restees in New York and Washington in- dicate that about one-half have positive urine specimens, but only about 20 per- cent have tests positive for heroin (mor- phine). In addition, preliminary findings from research in New York City indicate that arrestees with more than one drug in their urine (23 percent of all tested ar- restees) at the current arrest have had more arrests over the prior 3 years than persons with only one drug in their urine. Nevertheless, the percentage of arrestees with a positive urine test who are chronic, heavy drug users is still unknown. These issues must be resolved before urine test results can be used effectively to plan interventions for arrestees. Conclusion It has been established in studies of arrestees in two large cities that arrestees underreport their recent use of drugs when they are interviewed in criminal justice settings and that drug users cannot be identified readily from the charge at arrest. Urinalysis has been found to de- tect more drug use than persons admit to

72 in interviews held in criminal justice set- tings. In aciclition, there is some indica- tion that urinalysis soon after arrest may provide an indication of the risk of future criminal behavior Worst and Wish, 1983~. More research is needed to determine how best to use urinalysis tests to identify and process drug users. SUGGESTIONS FOR FUTURE RESEARCH In this section we present suggestions for future research on the relationship between drug abuse and crime. In the course of writing this paper and in re- sponding to the many questions raised by members of the Panel on Research on Criminal Careers, we have become aware of a number of gaps in the current knowI- edge about drug use anct crime. We dis- cuss below some of the questions that future research should strive to answer. When possible, we have inclicated possi- ble strategies for obtaining answers. Some of the areas that we have chosen to emphasize reflect biases that we wish to make explicit. In our opinion further elaboration and refinement of the "true" rates of crime among drug-using offenc3- ers should not be a primary goal of future research. Rather, we believe that future research should build on available find ings to determine ways to prevent the development of drug use and crime in persons at high risk for these behaviors, and to develop and assess the impact of interventions that may reduce these be- haviors in those who already exhibit them. In accordance with these priorities we recommence that the following ques- tions be addressed. What Is the Course of Drug Use and Crime in Persons Who Have Been Processed by the Criminal Justice System? Many studies have investigated how drug abuse treatment affects subsequent CRIMINAL CAREERS AND CAREER CRIMINALS drug use and criminal behavior. Far fewer studies have measured the impact of such criminal justice interventions as proba- tion, parole, pretrial cliversion, and incar- ceration on arrestees' drug use and associated crime. Studies are needec! of the impact ofthese conditions, both while the person is in custody or uncler super- vision and after the intervention period has enclecI. Do these forms of interven- tion permanently reduce drug use and associated crime or merely postpone them until the person can make up for lost time ? We know of no studies of the natural course of drug use and crime among ar- restees. Studies have found that older ar- restees are less likely to be using drugs near the time of arrest than are younger arresters. Have these older arrestees re- clucecl their abuse of drugs, perhaps be- cause of criminal justice interventions or drug abuse treatment? Or have the drug abusers merely matured out of criminal activities or dropped out of the arres tee population because of death or institu- tionalization? A long-term, prospective sturly of a cohort of offenders could an- swer these questions. A quicker and less expensive strategy woul(1 be to select a sample of persons arrester! in the past and contact each for a research interview; To- cating and reinterviewing an adequate sample of such persons might be clifficult, however. More information is needed about the types of interventions that are most effec- tive with clrug-involvecl offenders. Exper- iments with random assignment of per- sons to specific interventions shouIc! be conducted. For example, one study is cur- rently randomly assigning drug-involved persons to drug abuse treatment, to urine surveillance, or to a control group cluring the pretrial release period (Toborg, 1984~. The (differences in pretrial rates of rear- rest and failure to appear will be mea- sure(1 an(1 compared for each group. New types of supervised interventions shoul(1 also be tried. Studies of the Cali- fomia Civil Addict Program (McGIothTin,

TlIE IMPACT OF SUBSTANCE ABUSE ON CRIMINAL CAREERS Anglin, and Wilson, 1977) indicate that treatment accompanied by supervision was very effective while persons re- mained in the program. And a large stucly of treatment outcomes (Collins and Al- lison, 1983) has reported that persons sent by the court to treatment remained in treatment longer than persons referrer! by other sources. The traclitional view that persons have to enter treatment voluntar- iTy for it to be effective may be inaccurate. Court-orclerecI interventions accompa- nied by varying levels of supervision and sanctions should be tested. Interventions may have to be tailored to the characteristics of the particular of- fencler. Some studies have indicated that younger arrestees may have less severe drug-use patterns than older arresters. Younger offenders may therefore require interventions designed to cleter them from more serious involvement with drugs. Older arrestees with a long history of drug dependence may require treat- ment that maintains them on a drug such as methadone. A coorclinated intervention strategy that integrates criminal justice sanctions and treatment should be tried. The past practice of placing persons in the hands of criminal justice staff or treatment staff has been inadequate. Programs should offer all the services needed to address the diversity of life problems that offenders who use hard drugs invariably have (Hunt, Lipton, and Spunt, 19841. In a(lcli- tion, special programs that identify and treat the highest risk persons (e.g., violent predators) must be established and eval- uatec3. Successful treatment or incapacita- tion of these persons may produce the greatest impact on crime rates. What Impact Does Reducing the Availability of Drugs Have on Drug Use and Crimes? Behind the government's attempts to reduce the production and importation of illicit drugs is the assumption that these 73 efforts will raise the price ofthe drugs and thereby deter or reduce their use. The evidence for this is scanty at best and relies primarily on correlations of commu- nity-leve] indicators of drug use and crime. It is possible that higher drug prices may lead persons to commit an increased number of income-generating crimes. Given the cost of attempting to limit the supply of drugs in this country, a series of experiments should be conclucted to determine the impact of such efforts on street-level users. Law enforcement agents could target a location for intensive efforts to recluce the supply of heroin or cocaine. The impact of these activities could then be assessed through a compre- hensive set of direct and indirect indica- tors of drug use and crime in that area. The price and purity of the drugs in the stucly area could be measured, and criminal activity could be monitored through rates of arrest and reported victimizations. In addition, researchers could interview street-level users about their drug use and crime during the study period an(1 assess whether the local enforcement efforts merely encouraged users to go to other neighborhoods to obtain drugs. Such a study could use a neighboring area as a control group or a design in which the same neighborhood is studied before and after the experiment. Such an experiment would give an indication of the effects of the current multimillion dollar efforts to reduce the supply of drugs in the country. What Is the Relationship Between Drug Use and Crime Among Females and Do Females Require Specific Types of Interventions? The overwhelming majority of studies of drug use and crime have concentrated on males. It is true that the criminal jus- tice system processes a much greater number of mates than females. However, the magnitude of the drug abuse problem among females, especially those with a

74 history of prostitution, may actually be greater than that found among males. Fe- maTe arrestees in a number of cities have been found to be more likely to be using drugs at arrest than are mates. In a pilot study of females arrested in New York City (Wish, Bracly, and Cua(lraclo, 1984) 84 percent of the females with a history of prostitution who provided a urine speci- men shortly after arrest had a positive urinalysis result (almost 80 percent were positive for cocaine). These females also reported a level of drug abuse and asso- ciated problems that exceeded those found for male arresters. Female ar- restees with no charge for, or history of, prostitution tended to have rates of recent drug use comparable to those found for the mate arresters. The types of studies that have already been conducted with male offenders and those that we recommend in this report shouIct be conducted with female offend- ers. The types of problems female of- fenders have and the types of interven- tions that are effective with them shouIc! be analyzed. What Types of Crimes and Associated Problems Are Common to More Affluent Drug Users? The emergence of cocaine as a com- monly used drug throughout American society raises questions about how mid- dle-cIass users will finance their drug use. Does the use of cocaine and the resulting contact with the illicit drug market in- crease the likelihood that these persons will use heroin and other drugs? Will it make them more susceptible to violence (both as perpetrators and victims)? Such questions can be answered by conducting interviews with samples of micicIle-cIass users who are identified in large-sample drug surveys. Studies of high-risk popu- lations, such as persons who are arrested or who contact service agencies or physi CRIMINAL CAREERS AND CAREER CRIMINALS clans for assistance, could also offer information on these issues, although such findings would be limited to the type of persons studied. Prow Can Offenders Who Are Using Drugs Be Identified? Thousands of persons are processed by the criminal justice system each year, yet systematic attempts to identify those who are serious drug users are rare. Although urinalysis has shown some promise as a means of detecting persons using hard drugs, more research must be done to determine how current methods of proc- essing and recording information about arrestees can be improver] to identify drug users and high-rate offenders. Stucl- ies of programs that divert drug users into treatment programs on the basis of a short interview with the arrestee have failed to test systematically the validity of their referral methods. AncI many questions remain concerning the proper use ant! administration of the increasingly popu- lar enzyme multipliecl immunoassay test REMITS) urinalysis techniques (Morgan, 19841. Systematic evaluations of the abiT- ity of self-reports, urine tests, and official records to identify clrug-using offenders shouIct be conductecl. Methods of admin- istering, conducting, and reporting urine test results must be tested and stan(lard- ized. Confusion in the linking of urine test results with the original donor, for example, couIcl produce disastrous results for the offender and destroy any value of the testing. Assuming That Drug-Using Offenders Can Be Identifie(l, Can This Information Predict Future Criminal Behavior? Can identifying drug users at arrest or while on probation or parole permit the differentiation of persons at high risk for

THE IMPACT OF SUBSTANCE ABUSE ON CRIMINAL CAREERS future offending? Studies in a few juris- clictions c30 indicate that drug users ap- pear more recidivistic than other offend- ers. However, these studies used methods of detection (arrest charge, re- port of the officer, or urine tests) that have been found to have questionable ability to identify drug users. In addition, the level and types of drugs user] by offend- ers vary from jurisdiction to jurisdiction. Prospective follow-up studies that cIas- sify offenders according to current drug use and monitor their future offending, rearrest, and abscondence are needed. These studies should be replicated in many jurisdictions and should be con- clucted with offenders identified at the point of arrest, probation, and parole. Special emphasis needs to be given to studying the drug use and self-reported crimes of such persons after they have been released (with and without legal supervision). Future research should also examine whether certain types of nonus- ers are especially good candidates for early release or more lenient treatment. Assuming That Drug Use by an Offender Is a Good Predictor of Future Criminal Behavior, How ShouIc! This Information Be Usecl in Processing Offenders? There has been considerable cliscus- sion regarding the use of an offencler's drug abuse history in making administra- tive decisions Weinberg, 19841. The Comprehensive Crime Control Act of 1984 authorizer! preventive detention for defendants charger! with certain federal drug offenses and clirectec3 federal judges to consider a variety of factors, including history of drug and alcohol abuse, in set- ting pretrial release conditions. A number of questions still need to be addressed concerning judicial decision making at the state and local level. Should a person be deprived of pretrial release because of 75 current drug use, even if he or she is charged with a nondrug crime? ShouIc3 bait be set higher? Are preventive cleten- tion statutes needed to permit the use of (lrug abuse information in setting pretrial release conditions? Should urine surveil- lance or entry into a treatment program be mandated for all drug users or only for particular types of offenders? Or shouIcI urine testing and manclated treatment op- tions be limited to adjudicated persons at the point of the presentence investigation or referral to probation? Research is needed into the views of policymakers, judges, attorneys, and the public on these issues. Aside from the constitutional is- sues, each jurisdiction may have to cle- cicle how to proceed. How Can We Prevent the Onset of Serious Drug Use and Crime? The questions above have focused on the many persons who have already de- veloped patterns of hard-(lrug use and criminal behavior. They are, perhaps, easier to identify, but harder to treat. Identifying youths who are at high risk of developing these behaviors is probably more difficult, but such youths would presumably be easier to treat. Longitudinal studies of high-risk youths should be conducted. Enough is known to isolate groups of youths con- taining high-risk persons. Why some of them will cease their deviant behaviors while others continue shouIc3 be exam- inecl. Such studies may not always pro- vide information generaTizable to the en- tire population. However, reduction in serious criminality probably means hav- ing an effective strategy for only 2 percent or less of the youth population, anyway. CONCLUDING COMMENTS Research into the role of substance abuse in crime has come a Tong way.

76 Research questions have been refined to focus on key relationships regarding the role of drugs in the criminal event ant! to document the diverse patterns of drug use and criminal behaviors among cirug- using offenders. Measures of crime ant! drug use have advanced to the point of classifying types of offenders, assessing their rates of specific offenses in defined time periods, and obtaining information on multiple drug and alcohol use. The re- search into these issues shouIc! continue. Policymakers must also be informed about current research findings so that policies anti solutions reflect the prob- lems and neecis of the current drug-abus- ing offender. A little more than one-half the arrestees stucliec! in Washington, D.C., and New York City hac! recently used an illicit drug. Such information pro- vicles an unusual opportunity to inter- vene in and treat drug use in the most criminally active population. Yet, with one exception (Washington, D.C.), no ju- risdiction in the United States routinely identifies recent drug use in arrestees by . . urlna. .ySlS. Research can inform policymakers about the effectiveness of possible inter- ventions for iclentifiect drug abusers. Strategies are needed for reducing drug abuse and, thereby, crime on a short-term and Tong-term basis. Selective incapacita- tion is one alternative that may be useful for the most serious ctrug-abusing offend- ers; however, eliminating a person's drug use and offending while he or she is de- tained is unlikely to lead to a Tong-term cure. We believe that the greatest chance of success will derive from a coorclinatec3 strategy that assesses each offencler's di- verse set of problems and prepares an integrated solution that ciraws on a range of available options that extends from mandatory treatment and rehabilitation (Kaplan, 1983), to urine surveillance, ciaiTy supervision, and incarceration. It is CRIMINAL CAREERS AND CAREER CRIMINALS the challenge of future research to estab- lish how such a strategy can best be implementecI. APPENDIX A Methodologic Problems in Studying Substance Abuse and Crime The study of illicit hard-drug use is plaguer! by many of the same metho- dologic difficulties that affect the study of other criminal behaviors. Among the most significant problems that must be dealt with are (1) the rarity of harc3-drug use in the general population; (2) the validity of self-reports of harc3-drug use; (3) the fact that hard-drug use is often episoclic; (4) the fact that heavy use of any one (lrug is usually accompanied by heavy use of other drugs, inclu(ling alco- hol; and (5) the fact that measurement problems may be compoun(led when both drug use and other criminal behav- iors are assessed. The ways in which these methodologic problems can affect the valiclity of sub- stance abuse research are discussed be- Tow. In Appendix B. these problems are also discusser! as they apply to the studies of drug use anct crime being reviewed. Rarity of Hard-Drug Use in the General Population Surveys of the general population tend to find that use of illicit drugs other than marijuana is relatively rare. One survey (O'Donnell et al., 1976) found that only 100 men in a random sample of 2,510 American men aged 20 to 25 in 1974 had used heroin 10 or more times. In addition, the National Youth Survey (Elliott and Huizinga, 1984), a national probability sample of 1,725 noninstitutionalizec3 youth aged 11 to 17 in 1976, found few users of cocaine or heroin in each of the

THE IMPACT OF SUBSTANCE ABUSE ON CRIMINAL CAREERS five waves of interviews from 1976 through 1980. For each year fewer than 1 percent of the sample reported any use of heroin. The prevalence of cocaine use was somewhat greater as the cohort ma- tured, ranging from 1 percent in the 1976 interview sample to 10 percent in the youths reinterviewed] in 1980 (Johnston, Bachman, and O'Malley, 1982; Elliott et al., 19831. Use of heroin, cocaine, and PCP, drugs often thought to be associated wit-in criminal behavior, is even less com- mon in the adult population when oIcler persons are included. Because serious criminal behavior is also relatively rare in the general population, studies of heavy drug use and serious crime must neces- sariTy be limited to populations of persons at especially high risk for these behaviors. For example, the National Youth Survey found that only between 1.5 anct 2 per- cent of the sample members were multi- ple illicit drug users and serious offenders in 1976 or 1980 (Elliott and Huizinga, 1984), yet this group reported a dispro- portionate share of the crime and drug use found in the whole sample (Johnson, Wish, and Huizinga, 19831. Thus, much of what is known about hard-drug use and serious crime neces- sarfly comes from studies of adult ar- restees or incarcerated persons or from studies of unapprehended persons re- cruited from communities in which drug use and crime are prevalent. Although such studies cannot provide accurate estimates of drug use and crime relation- ships in the general society, they do enable researchers to study sufficient numbers of criminally active persons to untangle some of the complex interrela- tionships between drug use and crime. Nevertheless, the nature of the special populations being stucliecT (e.g., arrestees, incarcerated persons, unapprehendec3 persons) must be considerecl when assess- ing to whom the stucly results may apply. 77 Given that the focus of the Panel on Research on Criminal Careers is on seri- ous crime and the implications of a selec- tive incapacitation policy, in this appen- ctix we stress studies of high-risk aclults and refer only briefly to studies of youth in the general population. Validating Drug Use Barring objective observations of an- other person's taking verified amounts of a known drug in a laboratory or in some other controlled environment, research- ers typically measure drug use by the person's self-reports. Cognizant of the fact that study respondents may seek to conceal or underreport their use of illicit drugs, researchers have sought to vaTiclate self-reports of recent drug use through urinalysis. By comparing the respon- clent's reports of drug use with the urinal- ysis results, one can at least verify the accuracy ofthe self-reports, assuming that the tests themselves are valicI. There are many reasons why urinalysis may fait to detect drugs that have been taken within the previous 24 to 48 hours, however, including (but not limiter] to) the fact that the quantity and purity of the substance user! and the time since inges- tion are usually not known. In addition, a person could accurately report taking a (lrug that was, in fact, not the drug that he thought he had purchased in the illicit marketplace. Recent work (Wish, Strug, et al., 1983) has indicated, however, that the newer, more sensitive urinalysis (known commercially as EMITS) can de- tect the use of street-quality heroin (actu- ally morphine, the metabolite of heroin, is detected), cocaine, and methadone with a high degree of accuracy. The older, widely used mass-screening techniques (thin-layer chromatography) often fail to detect use of illicit, street-quality cocaine and heroin (Wish, Brady, et al., 19841.

78 CRIMINAL CAREERS AND CAREER CRIMINALS Although researchers may find it help . ~ ~ ~Polyclrug Use Among Heavy fuT to test the vanity of seli-reports ot Drug Users recent drug use through urinalysis, they still face many problems in vali(lating self-reports of drug use. For example, urine tests do not tell us when in the test-sensitive period (approximately 24 to 48 hours for cocaine and heroin) the drug use occurred. One cannot use urine tests to validate reports of drug use several days prior to the collection of the urine specimen. Researchers typically settle for an indication of the veracity of the re sponclent's answers to questions about drug use from the vaTiclation of recent drug use (by urinalysis) and from consis tency checks within the interview ant! between the interview and recorc] infor mation obtained for the respondent. Episodic Nature of Hard-Drug Use It is commonly believed that the heavy user of hard drugs, especially of heroin, uses the drug every day. It has become increasingly clear, however, that heroin users go in and out of periods of use even w i t h o u t t r e a t m e n t i n t e r v e n t i o n ~ M c G ~ o t h lin, Anglin, and Wilson, 1977; Robins, 1979; Ball et al., 1981; Ball, Shaffer, and Nurco, 1983; Johnson et al., 19851. Stud ies of active, unapprehenclecl heroin using offenders in New York City (Iohn son, 1984; Johnson et al., 1985) tend to find much polydrug use and drug switch ing, depending on drug availability and the person's finances and preferences at the time of purchase. The implications of the episodic nature of drug use for studies of drug use and crime are significant. One must not label a person a drug user over an entire period because the person reports being an acIdict or heavy user at one time during that pe riod. Measurement of drug use on a daily or weekly basis is needed to relate runs of drug use to changes in criminal behavior (see McGIothlin, Anglin, ant! Wilson, 1977; Nurco, Cisin, anct Balter, 1981a-c). Ample evidence exists that the heavier the use of any one ([rug, the greater the likelihood of use of other drugs. One study (Robins, HeIzer, et al., 1980) found that, as the use of alcohol or heroin in- creased in a sample of veterans, so cTi(1 the number of other drugs used in the same 2-year period. This study reporter] that persons addicted to heroin used an aver- age of 10.4 other drugs (out of 20) in a 2-year period after returning to the United States from Vietnam, comparer! with 7.9 drugs for less regular heroin users. In fact, the authors suggested that knowing how many illicit drugs were used by a person may be as good an indicator of severity of use as knowing which drugs are used. Studies of popula- tions of heroin-using offenders have tended to confirm this high degree of polycirug use (Strug, Stevie, et al., 1984; Strug, Wish, et al., 1984~. Both self-reports and urinalysis have inclicated that heavy users of any illicit drug use a smorgasbord of drugs, including alcohol, PCP, cocaine, heroin, pills, and illicit methadone fre- quently on the same clay. The fact that heavy users of heroin anti, incleed, of any illicit psychoactive drug tend to use multiple drugs presents some clifficulties for the researcher studying the relationship between (lrug use and crime. It may be misleading to attribute the criminal behavior of a heroin user to the heroin when that person is probably us- ing a multitucle of ([rugs and alcohol. Studies of drug use and crime must there- fore obtain precise information about all substances being used and control for their clifferential impacts on crime. A good example of this approach appears in the study of veterans cites] above (Robins, HeIzer, et al., 1980~. That study compared the effects of regular use of heroin, am- phetamines, marijuana, and barbiturates on social acldustment, after controlling for

THE IMPACT OF S UB STANCE AB USE ON CRIMINAL CAREERS early predisposing factors as well as other c3 rug use. The authors concluded that the greater social disability found among us- ers of heroin than users of other drugs was probably attributable to the types of per- sons who use heroin in our society persons with the greatest predisposition (apparent in their youths) to social prob- lems. Both the use of other drugs and one's disposition toward deviant behav- iors must be taken into consideration when ascertaining the impact of a partic- ular drug on one's social adjustment or criminal behavior. Measuring Drug Use and Crime We inclicatec3 above that persons may tend to underreport their use of illicit drugs and that urine tests can help to detect instances of recent drug use. When one measures both drug use and criminal behaviors by self-reports, however, there is another potential problem. Let us as- sume that person X is a seasoner! drug- abusing offender (perhaps in his late twenties) who is relatively open about his involvement in illicit behaviors. Such a person might report considerable drug use ant] crime in a research interview. On the other hand, person Y may be younger and as criminally active as person X, but less willing to admit to deviant behaviors. For example, some evidence exists that youthful offenders in Harlem were less likely to admit that they were junkies than were older offenders (Anderson et al., 1984~. Assuming that a sample contains many persons like X and Y. we could find a strong relationship between rates of drug use and crime that would be artifactual, resulting only from the fact that persons willing to disclose one of these behaviors are likely to disclose the others. A similar problem could occur if a respondent tended to view a particular time period to be one of general activity. In such cases he might report a high level of both drug 79 use and crime as a result of this general- ized belief about his life at that time. These biases or distortions in self- reported behaviors could be expected to increase as the time period being recalled gets further away from the time of inter- view (Bachman and O'Malley, 19814. On the other hand, in our research we have found some indication of underreporting among persons asked to report their drug use in the prior 24 to 48 hours during a research interview held in potentially threatening criminal justice settings. Re- searchers shouIc3, therefore, attempt to test drug and crime associations based on self-reports by comparing them against other information that does not depend solely on self-reports. An example of this strategy is the evaluation of the California Civil Addict Program by McGIothTin, Anglin, and Wilson (1977), in which an association between self-reported reduc- tions in narcotics use and self-reported reductions in crime was verified by a reduction in recorded arrests cluring the same period. APPENDIX B Studies of Drug Abuse an(1 Crime In this appendix we review studies whose findings greatly influenced our discussion in the body of this paper. We provide a brief summary of the design, major findings, significance, and potential limitations of each stu(ly. Evaluation of the CaZfornia Civil Addict Program (:McGIothtin, Anglin, and Wilson, 1977J Sample: Studied 949 men committed to the CaTifomia Civil A(l(lict Program. Included admissions from 1962~1963, 1964, and 1970. Many sample members entered this program as an alternative to serving time for a crime. Primary Measures: Self-reports clur

80 ing personal interviews, official arrest rec- orcts, and urine specimens. Validity Checks: For criminal behav- ior, arrest records; for recent drug use, urinalysis. Study Design: Follow-up study; nat- ural experiment using a comparison group of persons who were released from program early because of a legal techni- cality; oversampled a group of treat- ment successes as another comparison group. Significance of Study: This study is known for its excellent design and execu- tion. Both drug use and crime in specific time periods were measured, and runs of narcotics use were identified. The study indicated that the Civil Afflict Program, which consists of inpatient and outpatient periods, was effective while the men were in the program and less effective after termination. Supervision coupled with drug testing (originally with naline and later with urine tests) producer! mod- erate reductions in narcotics use and nondrug crime while the men were in the program. Methadone appeared to have a similar beneficial impact. The study con- cludec3 that reductions in daily runs of narcotics use couIc3 produce significant reductions in criminal behavior. Potential Limitations: Sampled only mates, largely those convicted of crimes. Applies to persons living in California; impact of methadone use by respondents not totally controlled for in the post-1970 analyses. Analysis of Drugs and Crime Among Arrestees in the District of Columbia (Wish, KZampp, et al., 1980J Sample: Consists of 57,944 men ant] women arrested and acljuclicatecl in the Washington, D.C., Superior Court from 1973 to 1977 and a recidivism file contain- ing 19,277 arrest cases (over a 6-year pe- riocI) for a sample of 7,087 consecutive per CRIMINAL CAREERS AND CAREER CRIMINALS sons arrester! in an 8-month panel period in 197~1975. Primary Measures: Prosecutors' case- processing records (from PROMIS), bait and sentencing information from court records, urinalysis results, ant! drug abuse treatment records. Validity Checks: For criminal records, none; for urine test results, some compari- son with police officer's perception of ar- restee's involvement with narcotics. Study Design: Cross-section case files of prosecutor's case information and results of a urine test from a specimen taken at arrest were merged for each in- dividual and analyzed; in addition, a per- son-based file containing arrest cases, uri- nalysis results, and information on drug treatment for 7,087 persons was con- structed and analyzed. Significance of Study: This study showed that urine test results could iden- tify arrestees at high risk of rearrest in a 4-year follow-up period. Drug-using ar- restees in a 6-year period prior to and after the index arrest had higher rates of bait violations and income-generating crimes than nonusers and equivalent rates of arrests for violent crimes. Female arrestees were more likely to be detected to be using drugs at arrest than male arresters. The report also contains infor- mation about the type of victims chosen by drug users and their types of arrest. Potential Limitations: Study looked only at arrest records and obtained no self-reports of crimes committed. Time at risk was not controlled for, although sub- sequent analyses indicated that adjust- ment for time at risk did not alter study findings. Drug use was measured only by . . urlna ySlS. Varieties of Criminal Behavior (Chaiken and Chaiken, 1982) Sample: Study of 2,190 inmates in prison or jail in Michigan, California, and

THE IMPACT OF SUBSTANCE ABUSE ON CRIMINAL CAREERS Texas. Sample was selected to represent an incoming cohort in the institutions. Analyses were weighted, where neces- sary. Primary Measures: Group-adminis- tered questionnaires nonconfidential- to enable linking of information with of- ficial records. Validity Checks: Performed exten- sive internal and external validity checks. Official records were available only for respondents in prison, however. External validity tests of self-reports versus record information inclicatec3 that 23 percent had "bad ciata" on 31 percent or more of 14 indicators checked. Internal consistency checks showed that between 28 and 32 percent of the respondents had trouble understanding the clefinition of the 2-year period preceding the primary measure- ment period. Approximately 17 percent of the sample hac3 bacl data on 21 percent or more of the 27 indicators of internal qual- ity that were checked. No systematic re- lationships were fount] between the global indices of internal and external validity and personal characteristics or reported crime rates in the measurement period. Official records on drug involve- ment were so poor that checks of the self-reportec3 information on substance abuse were impossible. Study Design: One-time, retrospec- tive, self-administerec3 survey question- naire and available official records were used. Significance of Study: Study is pri- mariTy known for the clevelopment of a typology of offenders having different levels of offending rates and for measur- ing individual offending frequencies (As) from inmates' self-reports. Self-report in- formation, but not official records, was useful in discriminating high- and Tow- rate offenders. Potential Limitations: Used identifi- able group-ac3ministered questionnaires, rather than personal interviews. Re 81 spouse rates were 50 percent in CaTifor- nia and Michigan prisons, 66 percent in California and Michigan jails, and 82 per- cent in Texas prisons. Measures of drug use were few and simple: did not mea- sure use of marijuana, PCP, or LSD after age 18; did not measure cocaine use at all. Method of drug administration was not measured. Alcohol use was measured by only a single, yes-no question regarding whether the person drank alcohol heavily, got drunk often, or had a drinking problem. The findings may apply to a select population of offenders, given the Tow probability of incarceration. Respon- clents' clifficulty in differentiating prior time periods places analyses over these periods in doubt. Statistical significance levels were often reporte(1 rather than the actual finclings, which limits the reader's ability to assess the magnitude of the differences reportecl. Criminality Among Heroin Addicts in Baltimore (:Ball et al., 1981; Nurco, Cisin, and Balter, 1981 a,b,c; Ball, Shaffer, and Nurco, 1983) Sample: A random sample of 243 mate opiate addicts arrested or identified by the Baltimore police department be- tween 1952 and 1971. Sample was strati- fiec3 by race and time periocl. Primary Measures: Personal follow- up interviews an(1 police, juvenile, and FBI records. Validity Checks: Interview informa- tion was checke(1 against recor(1 informa- tion on (late of birth, narcotics use, incar- ceration an(1 conviction history, and juvenile clelinquency history. Stucly Design: Follow-up interviews with sample in 1973 and 1974. Significance of Study: Known for its typology of heroin addicts and for its fin(lings regarding the number of crime days during periods of heavy narcotics use and lesser use.

82 Potential Limitations: Early papers presented findings for blacks and whites separately, since the sample hac3 been stratified by race to produce nearly equal proportions of whites and blacks. (On original list 77 percent were blacks; in the final sample 57 percent were blacks.) However, research on crime days by Ball et al. (1981) anc3 Ball, Shaffer, anc3 Nurco (1983) pooled blacks ant] whites without weighting the sample back to its original ethnic composition. Given the many dif- ferences found between black and white respondents (Nurco, Cisin, and Balter, 1982), the pooling of all subjects limits the generaTizability of results. Because whites were less criminally active than blacks, the disproportionate number of whites in the sample wouIcl tend to lower the estimates of crime. Persons were interviewed as long as 20 years after they had been iclentified as drug involved by the Baltimore police. Other than verifying that most of these persons were using drugs at about the time they were placed on the list (Bonito, Nurco, anc3 Shaffer, 1976), few checks were macle to verify that the behavioral patterns recalled were accurate. This study may also be limited by biases in self-reported behaviors (discussed in Ap- pendix A), which could have produced a strong association between drug use and crime in certain periods as an artifact of the measures. While the number of crime clays per year is measured, the study does not report data from which As may be computed. In addition, the computation of crimes per clay is not straightforward, given ambiguity in the way the pertinent questionnaire items were worcled. National Youth Survey (Elliott ant] Huizinga, 1984) Sample: Consists of 1,725 youths se- lectecI as a representative sample of American youths aged 11 to 17 in 1976. Persons were reinterviewed annually CRIMINAL CAREERS AND CAREER CRIMINALS from 1977 to 1981 about their delinquent anc3 drug-using behaviors. Primary Measures: Self-reports from personal interviews about behaviors clur- ing a 12-month period. Also obtained rec- ords of arrests. Validity Checks: For criminal behav- ior, used arrest records; for recent (1rug use, used internal consistency checks. Study Design: Prospective longitudi- nal design; each year of birth cohort was treated as an indepenclent sample. Significance of Study: This is the largest ant] longest study of a national sample of youths that is available. It is recognize(1 for its design and execution. Its reports contain measures of crime rates (routinely reported) for various types of clelinquent behavior. It provides the best information available about de- linquency in a large, representative sam- ple of youths. Potential Limitations: The major lim- itation is the small number of youths reporting extensive clelinquency anc3 se- rious drug use in a national probability sample. Heroin use was almost nonexis- tent (1 percent or less). While (dropouts from the study do not appear to slider substantially from reinterviewees, the Toss of even three to five highly clelin- quent youths could have re(lucecl crime rates. Economic Behavior of Street Opiate Users (Johnson et al., 1985) Sample: Consists of 201 heroin and methadone users recruite(1 from the streets of East and Central Harlem in New York City. Subjects were inter- viewed nine or more times an(1 provided a total of 11,400 person-days of informa- tion about their behaviors. Researchers were not successful in (leveloping a sam- pling frame from which to select persons with a known probability of selection. Primary Measures: Self-reports Bring 33 or more clays per person with

THE IMPACT OF SUBSTANCE ABUSE ON CRIMINAL CAREERS respect to crimes (type, number, clolIar, and drug income), drug patterns (use, purchase, sales, noncriminal income, noncIrug expenditures, arrests, and drug treatment. Validity Checks: For criminal behav- ior, some observations of subjects com- mitting crime in the streets, internal consistency checks cluring follow-up in- terview, reporting of similar crimes dur- ing different interviews; occasionally, two or more subjects reported about the same crime event. No arrest records were obtained. Reports of drug use were not validated, although many subjects report- ing use appeared intoxicated at the time of interview. Study Design: Convenience sample of persons encountered on the street who were screened by exaddict field-workers for heroin and methadone use and prob- able criminal behavior; subjects selected to represent the diversity of drug and criminal life-styles in the neighborhood; subjects reported to storefront research once each clay for 5 days and then 1 day per week (for 1 month or longer) to re- count prior week's activities. Significance of Study: This is the first study to compute crime rates from data about self-reported crime for persons while they were active on the streets. It is one of the few studies to present detailed daily and annualized data on the dollar returns from drug use, drug-ctistribution activities, and other crimes, as well as other economic behaviors. The study presents both quantitative and qualitative information about heroin abusers who are serious and regular criminal offenders. Potential Limitations: The sample is small and limited in geographical area, and it clid not follow accepted sampling procedures. It is unknown how represen- tative the respondents are of other of- fenders in New York City or in other cities. The As include numerous small drug-distribution crimes. The analyses seldom control for the effects of demo 83 graphic characteristics (sex, age, ethnic- ity, age of onset, education, etc.) on the As computed. Studies from the National Institute of ;/ustice-Funded Interdisciplinary Research Center (IRC) for the Study of the Relationship of Drug Use anct Crime (Strug, Stevie, et al., 1984; Strug, Wish, et al., 1984; Glassner et al., 1985) Samples: Three samples: youths in a moclerate-size city in New York (N = 1001; unapprehen~lec3, clrug-using aclult criminals in East Harlem (N = 179~; and aclults arrested for possession or sale of drugs in the East Harlem area (N = 116~. Primary Measures: Intensive, open- en(le(1 interviews of youths; structured personal interviews and urine tests for apprehended and unapprehended adult . . . criminals. Validity Checks: Internal checks and some corroboration by other youths; stud- ies of adults checked urine tests against self-reports of recent use of illicit drugs and found considerable concordance. Study Design: Studies of youths in- volvec3 three subsamples: a random sam- ple from school lists, a purposive sample based on field observations of deviant youths, and a sample of juveniles adjudi- cated as delinquent and residing in group homes or detention centers. Youths were interviewed for an average of 4 hours about their drug use, adjustment, and cle- v~ance. Studies of unapprehended and appre- hended adults in East Harlem: unap- prehended har(l-(lrug users who hac3 re- cently committed a serious nondrug crime were recruited from the streets en cl interviewed about the crime event anti the rote of drug and alcohol use in that event. A comparison group of 116 ar- restees were interviewed in a police sta- tion (in the same neighborhood as the unapprehended (lrug users ) about their

84 cTrug-use histories and a urine specimen was obtained. A second follow-up inter- view was obtained for 43 of these 116 persons after release. The releasees who showed up for reinterview were more indigent and reported more lifetime cle- penclence on drugs than those who dic3 not show up. Their educational and eth- nic backgrounds were similar to those who were not reinterviewed, as was their level of recent use of drugs. The fol- Tow-up interview was identical to the in- terview used with unapprehended per- sons and was followed by collection of a . . urine specimen. Significance of Study: Study of youths obtained in-clepth information about drugs and crime and successfully oversampled high-risk youths. Studies of aclult criminals obtained information about the role of drug and alcohol use in the crime event and examined the criminal justice system's processing of drug-involvec! arrestees. Potential Limitations: Study of youths had a small sample and collected much information in a qualitative way that limits quantification and extrapolation to other populations. The studies of unappre- hencled persons used paid recruiters to fincl persons who tract just committed crimes. The degree to which the stucly respondents are representative of other adctictect o~end- ers is unknown. The comparison group of arrestees does provide some indication of the potential biases in the data from unap- prehended respondents. The arrestees in the study were primarily arrested for pur- chase, sale, or possession of cocaine or her- oin, and some findings may not apply to arrestees charged with nondrug crimes or to persons arrested in over jurisdictions in New York City. Adclitional Current Studies Two ongoing studies of urine testing of arrestees, in Washington, D.C., and in New York City, have also influenced the CRIMINAL CAREERS AND CAREER CRIMINALS conclusions presented in this paper. The study in Washington (Toborg, 1984) is examining whether it is effective for judges to assign cirug-using arrestees to specific pretrial release conditions (treat- ment and/or urine monitoring) based on the results of a test of a urine specimen obtained shortly after arrest. The study in New York City (Wish, Bracly, and Cuadra- do, 1984, l9X5; Wish, Bracly, et al., 19~34; Wish, Chedekel, et al., 19~35) is examining the feasibility of using urine tests to iclen- tify arrestees at high risk of pretrial arrest and failure to appear in court. REFERENCES Anderson, Kevin, Wish, Eric, Johnson, Bruce D., Sears, Alton, and Miller, Tom 1984 Living Hard in the City: A Portrait of Ten Young "Hustlers." Paper presented at a meeting of the Academy of Criminal Justice Sciences, Chicago, Illinois, March. Anglin, M. Douglas, and McGlothlin, William H. 1984 Outcome of narcotic addict treatment in Cali- fornia. Pp. 10~128 in Frank M. Tims and Jac- queline P. Ludford, eds., Drug Abuse Treat- ment Evaluation: Strategies, Progress, and Prospects. Research Monograph 51. Rock- ville, Md.: National Institute on Drug Abuse. Bachman, Jerald G., and O'Malley, Patrick M. 1981 When four months equal a year: inconsisten- cies in student reports of drug use. Public Opinion Quarterly 45:536 548. Ball, John C., Shaffer, John W., and Nurco, David N. 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. Ball, John C., Roxen, Lawrence, Flueck, John A., and Nurco, David N. 1981 The criminality of heroin addicts when ad- dicted and when off opiates. Pp. 39 66 in James A. Inciardi, ea., The Drugs-Crime Connection. Beverly Hills, Calif.: Sage Pub- lications. Barton, William I. 1976 Heroin use and criminality: survey of state correctional facilities, January 1974. Pp. 419- 440 in National Institute on Drug Abuse and Research Triangle Institute, Drug Use and Crime: Report of the Panel on Drug Use and Criminal Behavior. Appendix. Research Tri

THE IMPACT OF SUBSTANCE ABUSE ON CRIMINAL CAREERS ~5 angle Park, N.C.: Research Triangle Insti- tute. Baxley, Robert C. 1980 Voluntary intoxication from phencyclidine: will it raise a reasonable doubt of the mental capacity of a person charged with a crime re- quiring specific intent or mental state?Jour- nal of Psychedelic Drugs 12(3-4):33~335. Bonito, Arthur J., Nurco, David N., and Shaffer, John W. 1976 The veridicality of addicts' self-reports in social research. International Journal of Ad- dictions 11(5):719-724. Bureau of Justice Statistics 1983a Prisoners and Alcohol. Washington, D.C.: U.S. Department of Justice. 1983b Prisoners and Drugs. Washington, D.C.: U.S. Depa~llent of Justice. Chaiken, Jan, and Chaiken, Marcia 1982 Varieties of Criminal Behavior. Santa Monica, Calif.: Rand Corporation. Chaiken, Marcia 1983 Crime Rates and Substance Abuse Among Types of Offenders. Unpublished paper. In- terdisciplinary Research Center, New York City. Clayton, Richard R., and Voss, Harwin L. 1981 Young Men and Drugs in Manhattan: A Causal Analysis. Research Monograph 39. Rockville, Md.: National Institute on Drug Abuse. Cohen, Jacqueline 1978 The incapacitative effect of imprisonment: a critical review of the literature. Pp. 187-243 in Alfred Blumstein, Jacqueline Cohen, and Daniel Nagin, eds., Deterrence and Incapac- itation: Estimating the Effects of Criminal Sanctions on Crime Rates. Washington, D.C.: National Academy Press. Collins, fames J., and Allison, Margret 1983 Legal coercion and retention in drug abuse treatment. Hospital and Community Psychi- atry 14(12~:1145-1149. Collins, James J., Hubbard, Robert, and Rachal, J. Valley 1985 Expensive drug use and illegal income: a test of explanatory hypotheses. Criminology 23:743-764. Collins, James J., Rachal, J. Valley, Hubbard, Rob- ert, Cavanaugh, Elizabeth R., Craddock, S. Gail, and Kristiansen, Patricia L. 1982a Crime and Crime Indicators in the Treat- ment Outcome Prospective Study. Research Triangle Park, N.C.: Research Triangle Institute. 1982b Criminality in a Drug Treatment Sample: Measurement Issues and Initial Findings. Research Triangle Park, N.C.: Research Tri- angle Institute. DuPont, Robert L., and Kozel, Nicholas J. 1976 Heroin Use and Crime. Paper presented at a meeting of the American Psychiatric Associ- ation, Miami Beach, Florida. Elliott, Delbert S., and Huizinga, David 1984 The Relationship Between Delinquent Be- havior and ADM Problems. Boulder, Colo.: Behavioral Research Institute. Elliott, Delbert S., Ageton, Suzanne S., Huizinga, David, Knowles, Brian, and Canter, Rachelle T. 1983 The Prevalence and Incidence of Delinquent Behavior: 107~1980. National Youth Sur- vey Report 26. Boulder, Colo.: Behavioral Research Institute. Fauman, Beverly J., and Fauman, Michael A. 1982 Phencyclidine abuse and crime: a psychiat- ric perspective. Bulletin of the American Academy of Psychiatry and the Law 10~3~:171-176. Feinberg, Kenneth R. 1984 Selective incapacitation and the effort to im- prove the fairness of existing sentencing practices. Unpublished paper. Kaye, Scholer, Fierrnan, Hays, and Handler, Wash- ington, D.C. Law and Society. Forst, Brian, and Wish, Eric 1983 Drug use and crime: providing a missing link. Pp. 84-95 in Kenneth R. Feinberg, ea., Violent Crime in America. Washington, D.C.: National Policy Exchange. Gandossy, Robert P., Williams, Jay R., Cohen, Jo, and Harwood, Henrick J. 1980 Drugs and Crime: A Survey and Analysis of the Literature. National Institute of Justice. Washington, 13.C.: U.S. Government Print- ing Office. Glassner, Barry, Loughlin, Julia, Johnson, Bruce D., Carpenter, Cheryl, Ksander, Margret, Berg, Bruce, and Stuck, Mary 1985 Drugs and Alcohol in Adolescent Delin- quency. Interdisciplinary Research Center. New York City. Goldstein, Paul J. 1981 Getting over: economic alternatives to pred- atory crime among street drug users. Pp. 67~4 in James A. Inciardi, ea., The Drugs- Crime Connection. Beverly Hills, Calif.: Sage Publications. 1985 The drugs/violence nexus: a tripartite con- ceptual framework. Journal of Drug Issues 15:493-506. 1983 Drug Related Involvement in Violent Epi- sodes. Proposal to the National Institute on Drug Abuse. New York City: Narcotic and Drug Research, Inc.

86 Greenberg, Stephanie W., and Adler, Freda 1974 Crime and addiction: an empirical analysis of the literature, 192~1973. Contemporary Drug Problems 3:221-270. Haberman, Paul, and Baden, Michael 1978 Alcohol, Other Drugs and Violet Death. New York: Oxford University Press. Heffernan, Ronald, Martin, John M., and Romano, Anne T. 1982 Homicides related to drug trafficking. Fed- eral Probation September:3-7. Huizinga, David, and Elliott, Delbert S. 1981 A Longitudinal Study of Drug Use and De- linquency in a National Sample of Youth: An Assessment of Causal Order. Boulder, Colo.: Behavioral Research Institute. Hunt, Dana, Lipton, Douglas S., and Spunt, Barry 1984 Patterns of criminal activity among metha- done clients and current narcotics users not in treatment. Journal of Drug Issues 14:687- 702. Inciardi, James A. 1979 Heroin use and street crime. Crime and Delinquency July:33~346. 1981 The Drugs-Crime Connection. Beverly Hills, Calif.: Sage Publications. 1984 Criminal Justice. Orlando, Fla.: Academic Press. Jessor, Richard 1979 Marihuana: a review of recent psychosocial research. Pp. 337-355 in Robert L. Dupont, Avram Goldstein, and John O'Donnell, eds., Handbook on Drug Abuse. Rockville, Md.: National Institute on Drug Abuse. Jessor, Richard, and Jessor, Shirley 1977 Problem Behavior and Psychosocial Devel- opment: A Longitudinal Study of Youth. New York: Academic Press. Johnson, Bruce D. 1973 Marijuana Users and Drug Subcultures. New York: John Wiley & Sons. 1984 Empirical patterns of heroin consumption among selected street heroin users. Pp. 101-123 in George Serban, eel., Social and Medical Aspects of Drug Abuse. Jamaica, N.Y.: Spectrum Publications. Johnson, Bruce D., Wish, Eric, and Huizinga, David 1983 The Concentration of Delinquent Offend- ing: The Contribution of Serious Drug Involvement to High Rate Delinquency. Pa- per presented at a meeting of the American Society of Criminology, Denver, Novem- ber. Johnson, Bruce D., Goldstein, Paul, Preble, Edward, Schmeidler, James, Lipton, Douglas S., Spunt, Barry, and Miller, Thomas 1985 Taking Care of Business: The Economics of CRIMINAL CAREERS AND CAREER CRIMINALS Crime by Heroin Abusers. Lexington, Mass.: Lexington Books. Johnston, Lloyd, Bachman, Jerald G., and O'Malley, Patrick 1982 Student Drug Use, Attitudes, and Beliefs: 1975~1982. Rockville, Md.: National Insti- tute on Drug Abuse. Kandel, Denise B. 1978 Longitudinal Research on Drug Use. New York: Halsted Press. 1984 Marijuana users in young adulthood. Ar- chives of General Psychiatry 41:20~209. Kaplan, John 1983 The Hardest Drug: Heroin and Drug Policy. Chicago, Ill.: University of Chicago Press. Kleinman, Paula H., Wish, Eric D., Deren, Sherry, and Rainone, Gregory 1984 The "Pure" Marijuana-using Client: Where? Unpublished paper. Narcotic and Drug Re- search, Inc., New York. Kozel, Nicholas J., ar~d DuPont, Robert 1977 Criminal Charges and Drug Use Patterns of Arrestees in the District of Columbia. Rockville, Md.: National Institute on Drug Abuse. LuPoff, Irving F. 1974 Issues in the evaluation of heroin treatment. Pp. 12~157 in Eric Josephson and Eleanor C. Carroll, eds., Drug Use: Epidemiological and Sociological Approaches. New York: John Wiley & Sons. 1976 Consequences of use: heroin and other nar- cotics. Pp. 19~227 in Joan Rittenhouse, ea., The Epidemiology of Heroin and OtherNar- cotics. Research Monograph 16. Rockville, Md.: National Institute on Drug Abuse. McBride, Duane C. 1983 Homicides and Violence among Miami Drug Dealers. Paper presented at a meeting of the Society for the Study of Social Prob- lems, Detroit, Michigan, August. McGlothlin, William H. 1979 Drugs and crime. Pp. 357~65 in Robert L. Dupont, Avram Goldstein, and John O'Don- nell, eds., Handbook on Drug Abuse. Rockville, Md.: National Institute on Drug Abuse. McGlothlin, William H., Anglin, M. Douglas, and Wilson, Bruce D. 1977 An Evaluation of the California Civil Addict Program. Services Research Issues Series. Rockville, Md.: National Institute on Drug Abuse. Morgan, John P. 1984 Problems of mass screening for misused drugs. Journal of Psychoactive Drugs 16(4):30~317.

THE IMPACT OF SUBSTANCE ABUSE ON CRIMINAL CAREERS Nurco, David N., Cisin, Ira H., and Baiter, Mitchell B. 1981a Addict careers. I. A new typology. Interna- tional Journal of the Addictions 16(8): 130~1325. 1981b Addict careers. II. The first ten years. Inter- national Journal of the Addictions 16(8): 1327-1356. 1981c Addict careers. III. Trends across time. Inter- national Journal of the Addictions 16(8): 1357-1372. 1982 Trends in the age of onset of narcotic addic- tion. Chemical Dependencies: Behavorial and Biomedical I ssues 4(3) :221-228. O'Donnell, John A., and Clayton, Richard R. 1981 The steppingstone hypothesis: a reappraisal. Chemical Dependencies 4. O'Donnell, John A., Voss, Harwin L., Clayton, Rich ard, Slahn' Gerald T., and Room, Rol~in G. 1976 Young Men and Drugs-A Nationwide Sur- vey. Research Monograph 5. Rockville, Md.: National Institute on Drug Abuse. Peterson, Robert C., and Stillman, Richard C., eds. 1978 Phencyclidine (PCPJ Abuse: An Appraisal. Research Monograph 21. Rockville, Md.: National Institute on Drug Abuse. Reuter, Peter 1984 The (continued) vitality of mythical num- bers. Public Interest 75: 13~147. Robins, Lee N. 1979 Addict careers. Pp. 325 336 in Robert L. Dupont, Avrarn Goldstein, and John O'Don- nell, eds., Handbook on Drug Abuse. Rock- ville, Md.: National Institute on DrugAbuse. Robins, Lee N., and Wish, Eric 1977 Development of childhood deviance: a study of 223 urban black men from birth to 18. Pp. 448-473 in Mae F. McMillian and Serrgio Henao, eds., Child Psychiatry Treatment and Research. New York: Brunner/Mazel. Also published in Social Forces 56(2):448-473. Robins, Lee N., Davis, Darlene H., and Wish, Eric D. 1977 Detecting predictors of rare events: demo- graphic;, family, and personal deviance as predictors of stages in the progression to- wards narcotic addiction. Pp.379-406 in J. S. Strauss, H. M. Babigian, and M. Hoff, eds., The Origins and Course of Psychopathol- ogy. New York: Plenum. Robins, Lee N., Helzer, John E., Hesselbrock, Michie, and Wish, Eric D. 1980 Vietnam veterans three years after Vietnam: how our study changed our view of heroin. In L. Brill and C. Winick, eds., Yearbook of Substance Abuse. New York: Human Sci- ences Press. 87 Robins, Lee N., Hessellbroc>k, Mic~hie, Wish, Eric D., and Helzer, John E. 1977 Polydrug and alcohol use by veterans and nonveterans. Pp. 74-90 in David E. Smith, Steven M. Anderson, Millicent Buxton, Nancv Gottlieb William Harvev and 7 ~ 7 Tommy Chung, eds., A Multicultural View of Drug Abuse: Proceedings of the National DrugAbuse Conference, 1977. Philadelphia, Pa.: Hall and Schenkman. 1~978 Alcohol and Crime in Veterans. Paper pre- sented at the National Institute on Law En- forcement and Criminal Justice Colloquium, Washington, D.C. Sells, Saul B., Stimson, D. D., Joe, G. H., Demaree, R. G., Savage, L. J., and Lloyd, M. R. 1976 A national follow-up study to evaluate the effectiveness of drug, abuse treatment: a re- port on cohort I of DARP five years later. American Journal of Drug and Alcohol Abuse 3(4~:54~556. Siegel, Ronald K. 1978 Phencyclidine, criminal behavior, and the defense of diminished capacity. Pp. 272 288 in Robert C. Peterson and Richard C. Stillman, eds., Phencyclidine (PCP) Abuse: An Appraisal. Research Monograph 21. Rock- ville, Md.: National Institute on Drug Abuse. 1980 PCP and violent crime: the people vs. peace. Journal of Psychedelic Drugs 12(3~):317- 330. Silverman, Lester P., and Spruill, Nancy L. 1977 Urban crime and the puce of heroin. Journal of Urban Economics 4:8(}103. Simonds, John F., and Kashani, Javad 1980 Specific drug use and violence in delinquent boys. AmericanJournal of Drug and Alcohol Abuse 7(3&4~:305 322. Singer, Max 1971 The vitality of mythical numbers. Public In- terest 23:~9. Single, Eric, and Kandel, Denise B. 1978 The role of buying and selling in illicit drug use. Pp. 118-128 in Arnold Trebach, ea., Drugs, Crime, and Politics. New York: Prae t~er. Strug, David, Stevie, B., Wish, Eric, Johnson, Bruce D., Miller, Tom, and Anderson, Kevin 1984 Hustling to Survive: The Role of Drugs, Alcohol, and Crime in the Lid of Street Hustlers. Paper presented at a meeting of the Academy of Criminal Justice Sciences, Chi- cago, March. Strug, David, Wish, Eric, Johnson, Bruce D., Anderson, Kevin, and Miller, Tom 1984 The role of alcohol in the crimes of heroin

88 abusers. Crime and Delinquency 30(4):551- 567. Tinklenberg, Jared R. 1973 Drugs and crime. Pp. 242~299 in National Commission on Marihuana and Drug Abuse, Drug Use in America: Problem in Perspec- tive. Vol. I, Appendix. Washington, D.C.: U.S. Government Printing Office. Tinklenberg, Jared R., and Ochberg, Frank M. 1981 Patterns of adolescent violence: a California sample. Pp. 121-140 in Jared R. Tinklen- berg and Frank M. Ochberg, eds., Behavioral Aspects of Aggression. New York: Alan Liss. Tinklenberg, Jared R., Roth, W. T., Kopell, B. S., and Murphy, P. 1976 Cannabis and alcohol effects on assaultive- ness in adolescent delinquents. Pp. 85-94 in Chronic Cannabis Use. Annals of the New York Academy of Sciences 282. Tinklenberg, Jared R., Murphy, Peggy, Murphy, Patricia L., and Pfefferbaum, Arnold 1981 Drugs and criminal assaults by adolescents: a replication study. Journal of Psychoactive Drugs 1343~:277-287. Toborg, Mary 1984 Preliminary Findings on the D.C. Urine Testing Experiment. Paper presented at a Meeting of the American Society of Crirni- nology, Cincinnati. Voss, EIarwin L. 1976 Young men, drugs and crime. Pp.351-385 in Robert Shellow, ea., Drug Use andCrime: Appendix, Report of the Panel on Drug Use and Criminal Behavior. Research Triangle Park, N.C.: Research Triangle Institute. Washton, Arnold M., and Gold, Mark S. 1984 Chronic cocaine abuse: evidence for adverse effects on health and functioning. Psychiat- ricAnnals 14(10):733-743. Weissman, James C. 1979 Understanding the drugs and crime connec- tion: a systematic examination of drugs and crime relationships. Journal of Psychedelic Drugs 10:171-192. Winick, Charles S. 1962 Maturing out of narcotic addiction. Bulletin on Narcotics 14:1-7. Wish, Eric D. 1982 Are [Ieroin Users lleally Nonviolent? Paper presented at a meeting of the Academy of Criminal Justice Sciences, Louisville, Ken- tucky, March. 1986 PCP and crime: just another illicit drug? Pp. 174-189 in D. PI. Clouet, ea., Phencyclidine: an Update. National Institute on Drug CRIMINAL CAREERS AND CAREER CRIMINALS Abuse. Research Monograph No. 64. Rock- ville, Md.: National Institute on Drug Abuse. Wish, Eric D., Brady, Elizabeth, and Cuadrado, Mary 1984 Female Arrestees: The Most Serious Drug- abusers? Paper presented at the annual meeting of the American Society of Crimi- nology, Cincinnati, Ohio, November. 1985 Drug Use and Crime in Arrestees in Manhat- tan. Presented at a poster session of the Forty-seventh Annual Scientific Meeting of the Committee on Problems of Drug Depen- dence, Baltimore, Maryland. Wish, Eric D., Anderson, Kevin, Miller, Tom, and Johnson, Bruce D. 1984 Drug Use and Abuse in Arrestees: New Findings from a Study of Arrestees in Man- hattan. Paper presented at a meeting of the American Criminal Justice Society, Chicago, Illinois, March. Wish, Eric D., Brady, Elizabeth, Cuadrado, Mary, and Sears, Alton 1984 Preliminary Findings from the "Drug Use as a Predictor of Pretrial Behavior in Arrestees" Project. Paper presented at the annual meet- ing of the American Society of Criminology, Cincinnati, Ohio, November. Wish, Eric D., Chedekel, Morris, Brady, Elizabeth, and Cuadrado, Mary 1985 Comparison of the Use of Thin-layer Chro- matography and EMITS for Detecting Re- cent Drug Use by Arrestees. Paper pre- sented at the annual meeting of the American Academy of Forensic Sciences, Las Vegas, Nevada. Wish, Eric D., Klumpp, K. A., Moorer, A. PI., and Brady, E. 1980 An Analysis of Drugs and Crime Among Arrestees in the District of Columbia. Springfield, Va: National Technical Informa- tion Service. 1981 Executive Summary: An Analysis of Drugs and Crime Among Arrestees in the District of Columbia. Washington, D.C.: National In- stitute on Justice. Wish, Eric D., Strug, David, Anderson, Kevin, Miller, Tom, and Johnson, Bruce D. 1983 Are Urine Tests Good Indicators of the Va- lidity of Self-Reports of Drug Use? It De- pends on the Test. Paper presented at the American Society of Criminology Meeting, Denver, Colorado, November. Zinberg, Norman 1984 Drugs, Set, and Setting. Chicago, Ill.: Uni- versity of Chicago Press.

Next: 3. The Relationship of Problem Drinking to Individual Offending Sequences »
Criminal Careers and "Career Criminals,": Volume II Get This Book
×
Buy Paperback | $110.00
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

Volume II takes an in-depth look at the various aspects of criminal careers, including the relationship of alcohol and drug abuse to criminal careers, co-offending influences on criminal careers, issues in the measurement of criminal careers, accuracy of prediction models, and ethical issues in the use of criminal career information in making decisions about offenders.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!