from the Bureau of the Census reveal that at the close of the 1970 census, the median family income of New York City residents was $9,692. while, the mean income of families reporting public assistance in that city was $2.114.

If we assume that a heroin-dependent person has a daily habit of $20 (the lowest daily estimate found but on(, atuountin(y to $7.300 per years), -xecanalsolISSUDWthat it Avouldbe relatively inipossible forhiiii to support his habit without supplenientin(y his Income throu(rh illegal means.

F or the most part, this Ille'-al activity manifests itse" in crimcs against property, principally burglary an@' shoplifting (Finestone, 1966; Kozel, et a]., 1969; Preble and Casey, 1969, Jacoby, et 11., 1973). Le@@F, coninion] ' v, licioln-dependont pers(IIS II)a - v resort, to offenses against the person (assault, IIIII(ro-'11g, rob

bory) In It similar- desperate effort to obtain the nionvy required to purchase the drug (Anisel, et aL, 1971; Friedman and Friedman, 1973; Jacobv, et aL, 1973; Tinklenberg. 1973). Researchers have estimated that for heroln-dependent person to support his habits lie must steal property amounting), to between two and one-half to file tinies the actual cost of his habit (Cushman, 1971) and that lie may be (expected to steal in property an amount ranging f rom $'2,5,000 to s,50.000 per year (Joseph and Dole, 1970; DuPont, 1.972). The (Estimated daily tolls in stolen property have ranged from $25 to $'375. At least one res(,archer (Singer, 1971), liowever, has been highly critical of these, estimates, all of which lie deems to b



e grossly exaggerated.

Heroin-dependent persons are also known to raise funds to support their habits by piniping and prostitution. Although these activities do not represent a direct financial burden to society, they (lo, nevertheless, constitute socially and legally disapproved behavior" (Cushman. 1971 , 1972; seealso Ainsel, et al., ini).

A lieroin-dependent person may also b(Tonie involved in the coinplicated hierarchy of individuals who I re responsible f or the, importation and distribution of heroin. In this way, lie is assured of a supply of the, dru(r foi- his own needs and verv often realizes a profit large enough to induce him to go into the dru(T-dealing business for himself.

Criminal Justice Costs: Processing the Drug-Dependent Person

In addition to the costs of crime incurred by drug-dependent persons, the community must also assume the cost of investigating. identifying, arresting, detaining, trying, sentencing, treating and rehabilitating thedrug-dependent offender.

In 1969, the New York State Department of Corrections estimated that it cost the State $12.75 per day for incarceration of one person

(Cushman, 1971), a figure including the separate costs of overhead, custodial personnel, maintenance and other limited Sel.ViCeS.15 A sim'lar

figure of $12 per day was quoted as the cost of (retention, while arrest procedures were estimated to total $25 per arrest (Cuslinian, 1971)."

In a more recent study of beroin use in Washington, D.C. (DuPont, 1972), the daily cost per user for incarceration was set at,$14, compared with an approximate cost of $5.50 per patient for outpatient care. The Corrections Department in thwart saine city listed the, daily cost per offender on parole at $0.97, a figure someNvIiat lo%ver than that noted for New York a few years ago.

If the costs of arrest,, trial, incarceration, treatmei)t and the like are multiplied by the alleged number of heroin-dependent, persons in this country today, society is faced with anotlier potentially astronomic bill directly related to heroin and otl)er dru- use and dependence. To this figure must be added additional amounts which reflect the, rate of recidivism among drug offenders and the costs Incurred froin Crimes committed to support their habits. Yet another adjustment must be made for those who, during the course of flieir drug dependerice. Avill probably be at-rested several times oil a varietv of char-es and be processed through the criminal justice system many times (veil.

t_1

The Effect of Drug Treatment on Crime

During the past eigrlit years, methadone maintenance has evolved from an experittiental prog-rani ivith 2-@ patients into tbe most Nvidely used treatment mortality in the nation. Its supporters ba\'e claimed that participation in Ole metbadone pro-rain reduces lierolii-related crime, a direct function of tlie patient's reduced need for -Valid use of heroin. III addition, iiidividuals participating III methadone maintenance prograinsare reportedtobe ableto finwhoii, perbaps for the first tinic in tbeir adult lives, as socially responsible aial ecoimmically independeiit and productive members of society.

The, Corim-iissioii has carefully scrutinized these claims. III (r,neral, they have been based onthe reported flii(iiiigs of a 11inited mimber of quasi-experimental studies (mostly before-after comparlsorls) ; Ole deficiencies in research design, samplillgr t(,(.Illll(illes"liltlvsistil(I interpretation of the data sigrnificantly ]Unit flie reliability '111d \-'Ilidity of

One 1969 study (Babst, et al., 1969) estimated that the cost of keeping all individual oil parole or in aftercare came to approximately one-tenth of that required to maintain him in a correctional facility. This suggests all a(Witional cost to the community of approximately $1.25 per day once the individual is released from prison (if. of course. the estimated cost of incarceration is reasonably accurate).

"This estimate was derived hy multiplying the hourly wag", of an average policeman by five, the absolute minimal estimated time deemed necessary to complete the initial administrative procedures attendant to arrest.

the conclusions and the inferences drawn therefroin. Tri fact, we have not, found sufficiently responsible research to conclude that any of the N,arious treatment modalities, regardless of type.actually reduce crime.

When metliadone inaintenance NA-as first dewised, its r1ine-reducing potential was regarded as ancillary to aclilex@einent of the primary pro-

gram goal, the elimination of livroiii use. With the passave of time,

however, heightened public fear and dramatic g

frequently reported to he heroin-related acquisitive (Laurie have distorted the public's view of the, purpose of inetliadone maintenance. Funding agencies, pro-ram officials and public leaders li,,ve contilillally touted this treatment modality as a law enforcenient strategy designed primarily to reduce drug-related crinie.11

Current Research Findings

The de-ree to which patients in treatment, have made gains toward crime-free lifestvies is generally measured by comparing patients' arrest records prior to pro-rani entry and during treatinent. In a few studies, however, attempts at control leave been made by con-1paring

e,

arrest records of methadone maintenance patients with those of reportedly niatelied groups of individuals in otlier treatment modalities (detoxification or abstinence, programs) or with those of individuals who dropped out of the, maintenance program at, some stage.

Several before-after comparison studies have attempted to establish a direct relationship between methadone maintenance, laid a reduction in crime. Although the data available strongly suggest a reduction in the criminal beliaN-ior of patients in treatment, differences in the basis of measurement preclude precise comparison of tbeir results. Joseph and Dole (1970), for example, utilized data on com-letions whereas Gearing (1971) studied arrests; Cushman (1971) reported on data derived f rom police records and interviews while DuPont (1972) drew on crimes known to the police, arrests and incarceration figures. The, periods of time before and after treatment which were measured also differ considerably and thereby affect the ernue rates reported.

"The Canadia n LeDain Commission recently noted in its, report on treatment "The success of methadone programs is generally measure(] in terms of social

or cultural criteria rather than psychological ones. A return to a normal life cycle based upon employment, marriage and stability of social interaction is the prime criterion of success. However, the expectation of a normal life style characterized by emotional maturity and complete abstinence in addition to a law-abiding. productive existence is an ideal which in practice is seldom achieved. Because of this, some methadone programs arrange their goals hierarchically. Tile foremost expectation is that all patients who are treated will become law-abiding citizens. although they might not become productive. mature or drug-free. The next level is to achieve a status in which patients are law-abiding and gainfully employed." (1972. 1). 24.)

in what is

Examination of studies incorporatim, control (rroups bave generally come to conclusions similar to those reflected in tbe before-after (lesigrns. Researchers who have conipared individuals in treatment with those who have dropped out, at some stage uniforinly show significant differences in arrest or conviction rates between the two groups and therefore conclude that it Avas the treatment which Nvas largely responsible for the resultant, crime reduction (Josepli and Dole, 1970, W11liams, et, at, 1970; Aloffett, et aL, 1971: ('lines, 1972, DuPont, 194-2).

Studies designed to compare tile arrest or conviction rates of Indl-

viditals in different treatment modalities are intended to deal more

ifically with the de(gee of (,Jim(, reduction produced by a partlC_ speci I

iflar type of treatment. In -eneral, their finding@. have shown that methadone maintenance has been significantlY niore siweessful than othermodalities (d(,toxifi(,@ttloii,,iii(lal)stliieii(@e.foi-exiiiil)](,) Inachieving crime reduction, althou.-Ii the absolute :amount of reduction and tile prograin differentials have varled considerabiv (Asher, 1970: Joseph and Dole, 1970; Wieland and Chambers, 1970; DuPont, 1972); CuskeY et al., 1973).

Although the statistics presented are (General]].\, Impressive (crime reduction estimates raiwe front 14% to untested basic issillilptions and other methodological deficiencies preclude their -eneraliz.1tion and sio-nal (Careful interpretation.

In one before-after shidy, for example, tile, researcher's major hypothesis was that a recent airline] its(, in serious, (,time lar-e1v attributable to a heroin "opidemic," and was manifest in signilficant increases in commitment rates of oplate offendersand parallel increases in index offenses for that period. A second hypothesis Nvas that a suhsequent 1111expectedand abrupt devillic, Ul S"lliolis (111111le IvIls largel vattriblitable to the treatment (priniaril.v methadone niaintenance) of thousands of licroin-dependent persons. Although the researcher did Z-M'e SMI)e credit for this "pronotinced find progressive" reduction in (,Jim(, to the doubling of the police force during the tinle period under investi-ation, the major conclusion of the StIldN, -was that if tile rise In crMle \VIIS correctlY attributable to a ]reason) epidenlic, their it was al@'o true their heroin treatment Nvas lar-ely responsible foi- the subsequent cl-IIII(I 1'eduction (Dul'ont, 1971).

The researcher failed to denionstrate, however, that the si-nificant increase in reported crime Ai-as indeed attributable to the alleged rise in heroin dependence. Althmigh the proportion of incarcerated heroindependent persons did increase. the reported number of index crillies increased I)v all even greater niar-In. '@;onie of ties increase was undoubtedIv related to the (,rowth of heroill dependence, but the very large and general increase in the crinie rate su-(rest that most of the increase was probably attributahle to tile inajority of crinlinals who

are not heroin -depen dent, Anotherf all acy in the researcher'sreasoning lies in the, fact that the studv was based on increases in arrests and incarceration and on the records of those persons who were in treatment. Those arrested, however, represent only a small fraction of the Universe of offenders, e@-cn drug offenders, persons incarcerated constitute onlv a tHiv friction of those arrested, and persons in treatment represent only a small propoition of known heroin-dependent persons (many more are on waiting lists) which, in turn, represpnts some unknown fraction of all heroin-dependent persons. Thus, what is perceived as a trhunph of a particular treatment modality may be for the most part a simple statistical artifacts

Numerons other methodological obstacles preclude generalization from research findings regarding, treatment an(] crime. In only one study, for example, did the researcher attempt to control for dosage level of the methadone administered (those with lower dosages were found to have higher erime rates). None of the studies (Examined takes into account, the monitoring and supervision of the drug's administration nor is there any standardization of the point in treatment at which the measurements are. made. Finally, some studies rise a man-year figure as a measurement of success although the conclusions drawn therefrom are (Especially likely to be distorted and misleading, (Joseph and Dole, 1970; Gearing, 1971, Joseph, 1972).

To illustrate this final point. let, its assun-w. for a moment, that we are interested in 20 individuals in treatment program X. Of thesis 20, 15 had dropped out at exactly two months after enter,, two more dropped out after file months in the program, one staved for a vear before leaving and one confirmed in treatment for three ' years. -Afultiplying the member of individuals (20) by the mimber of months each spent in the pro(yrain and addin,(-, those fi-ures vields the total member of man-inontlis spent in treatment. This member, divided by 12, equals the total number of man-years in treatment. (See Table