Estimating drug treatment needs among state prison inmates

Drug and Alcohol Dependence 77 (2005) 269–281

Estimating drug treatment needs among state prison inmates Steven Belenkoa,∗ , Jordon Peughb,1 a

Treatment Research Institute at the University of Pennsylvania, 600 Public Ledger Building, 150 South Independence Mall West, Philadelphia, PA 19106-3475, USA b Harris Interactive, 111 Fifth Avenue, New York, NY 10003, USA

Received 6 January 2004; received in revised form 18 August 2004; accepted 18 August 2004

Abstract Growing prison populations in the U.S. are largely due to drug-related crime and drug abuse. Yet, relatively few inmates receive treatment, existing interventions tend to be short-term or non-clinical, and better methods are needed to match drug-involved inmates to level of care. Using data from the 1997 Survey of Inmates in State Correctional Facilities, a nationally representative sample of 14,285 inmates from 275 state prisons, we present a framework for estimating their levels of treatment need. The framework is drawn partly from the American Society of Addiction Medicine Patient Placement Criteria and other client matching protocols, incorporating drug use severity, drug-related behavioral consequences, and other social and health problems. The results indicate high levels of drug involvement, but considerable variation in severity/recency of use and health and social consequences. We estimate that one-third of male and half of female inmates need residential treatment, but that half of male and one-third of female inmates may need no treatment or short-term interventions. Treatment capacity in state prisons is quite inadequate relative to need, and improvements in assessment, treatment matching, and inmate incentives are needed to conserve scarce treatment resources and facilitate inmate access to different levels of care. © 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Drug treatment; Offenders; Level of treatment need; Placement criteria

1. Introduction America’s prison and jail populations have grown substantially over the past 20 years. Between 1980 and 2002, the total number of inmates in the United States quadrupled from 501,886 to 2,033,331 (Harrison and Beck, 2003). The state prison population (housing inmates convicted of state crimes, usually felonies, who are serving sentences of more than a year) increased by 309% to 1,209,640, the federal prison population (housing inmates convicted of violating federal laws) increased by 538% to 151,618, and the local jail population (inmates convicted of violating state laws, generally misdemeanors, who are serving sentences of less than ∗ Corresponding author. Tel.: +1 215 399 0980x119; fax: +1 215 399 0987. E-mail addresses: [email protected] (S. Belenko), [email protected] (J. Peugh). 1 Tel.: +1 212 539 9706; fax: +1 212 539 9669.

0376-8716/$ – see front matter © 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.drugalcdep.2004.08.023

a year, or those held in custody while awaiting trial in state courts) increased by 265% to 665,475 (Harrison and Beck, 2003). These increases have been fueled mainly by drugrelated crime, and the consequent high rates of substance abuse or dependence among inmates (Belenko and Peugh, 1998; Blumstein and Beck, 1999). Despite this large and growing number of drug-involved inmates, relatively few receive treatment while incarcerated (Belenko and Peugh, 1999), and the available treatment opportunities often are a choice between interventions that may be too limited and short-term for many substance-involved inmates (12-step programs or drug education classes), or those that are overly intensive and expensive (long-term residential treatment). Although data are lacking on the effectiveness of short-term or “outpatient” correctional treatment, a number of studies have found that participation in residential treatment during incarceration, followed by continuing care in the community, yields reductions in recidivism and relapse to drug use (e.g. Knight et al., 1999; Martin et al., 1999).

270

S. Belenko, J. Peugh / Drug and Alcohol Dependence 77 (2005) 269–281

However, long-term residential treatment beds are limited in correctional systems, and such treatment is not an option for inmates facing relatively short incarceration terms, such as jail inmates or parole violators (Deitch et al., 2002). This is unfortunate because many of the large numbers of inmates who are reincarcerated for drug-related parole violations do not receive the treatment they need and are simply returned to community, with a high likelihood of relapse. Little is known about the number of inmates who need different amounts or types of treatment, in part because of the absence of standardized and validated clinical screening and assessment in correctional facilities (Knight et al., 2002). Coupled with limited resources and treatment space, correctional systems face difficult problems in allocating scarce treatment resources and matching the level of care to the treatment need. Given the high rates of relapse and recidivism among released inmates, however (Marlowe, 2003, Petersilia, 1999), it is important to improve systems for linking druginvolved inmates to the most appropriate levels of care. Creating procedures and guidelines for more effectively matching drug-involved inmates to different levels of treatment can be informed by the literature on treatment matching in the general substance-involved population. Awareness of the benefits of matching patient to treatment levels and modalities dates back some 15 years and was originally driven largely by efforts to contain costs in the wake of managed care (Gastfriend and McLellan, 1997) and limits in treatment capacity (Hser et al., 1999), issues that remain salient today. The Institute of Medicine reports on treating alcohol and drug problems (Institute of Medicine, 1990a, 1990b) recognized that the different levels of severity and types of drug and alcohol problems suggested the importance of matching client needs to treatment type and intensity. The notion of matching stems from the idea that no treatment is effective for all clients, but that all treatment is effective for some clients (Gastfriend and McLellan, 1997); determining which clients will do better in which settings is the challenge of matching. One of the earliest widely disseminated matching schema was known as the Cleveland Criteria, developed in the late 1980s (Gastfriend and McLellan, 1997) through a multi-agency consensus process. Spurred by the popularity of the Cleveland Criteria, the National Association of Addiction Treatment Providers, and the American Society of Addiction Medicine (ASAM) worked together to develop the first ASAM Patient Placement Criteria (Hoffmann et al., 1991). The ASAM criteria provide guidelines for placement of patients in a hierarchy of five treatment settings ranging from early intervention through intensive inpatient treatment (Mee-Lee et al., 2001). Despite the ensuing promulgation and popularity of the ASAM and other matching protocols, evidence of their predictive validity in terms of treatment outcomes is still limited (Gastfriend and McLellan, 1997; Melnick et al., 2001; Thornton et al., 1998; Turner et al., 1999). There are two key dimensions to the matching problem: the severity of drug use and the other service needs. Evi-

dence that clients with a higher severity of drug use have better outcomes in residential/inpatient or more intensive or highly structured treatment comes from the DATOS study (Simpson et al., 1999), studies in therapeutic communities (Melnick et al., 2001), outpatient settings (Rychtarik et al., 2000; Thornton et al., 1998), and Project MATCH for alcohol patients (Project MATCH, 1998). Mattson et al. (1994) reviewed 31 studies that supported the notion of treatment matching. In addition, a number of studies in different treatment settings have found that matching services to specific client needs (e.g. psychological services, housing, employment, etc.) improves treatment outcomes (Gastfriend and McLellan, 1997; Hser et al., 1999; McLellan et al., 1983, 1993; Mattson et al., 1994; Moos and Finney, 1995). However, in many treatment programs, specific client service needs are not being adequately met (Etheridge et al., 1995; Hser et al., 1999). The Client Matching Protocol recently developed for therapeutic community clients by Melnick et al. (2001) combines dimensions of prior drug use pattern and severity, social factors, and education and work skills in an algorithm designed to determine whether a patient should be place in outpatient or residential treatment. This experience suggests that determining treatment need for inmates is not a matter of simply assessing for drug abuse or dependence. Many inmates present with an array of health and social problems that accompany their substance abuse (Belenko and Peugh, 1999; Hammett et al., 1998). Poor education, lack of employment, physical and mental health problems, lack of housing, and family instability are common among inmates and can undermine treatment and recovery (Beck and Maruschak, 2001; Belenko et al., 2003; Ditton, 1999; Finn, 1999; Taxman, 1999). For example, given the connections among crime, poverty, and poor health, it is not surprising that many inmates enter prison in need of medical services (Anno, 1991; Hammett et al., 1999; Marquart et al., 1997). Health services of particular relevance for drug-involved inmates include mental health services and services for the treatment of HIV and other infectious diseases (Hammett et al., 1998). For drugusing women offenders, sexually transmitted disease treatment services and pre- and post-natal care are often needed (Peugh and Belenko, 1999). A number of studies have found high rates of co-morbid substance abuse and mental health conditions among inmates and other offenders (Belenko et al., 2003; Ditton, 1999; Lamb and Weinberger, 1998; Teplin, 1994; Teplin et al., 1996). Drug-involved inmates frequently have educational deficits and sporadic work histories, which can affect longterm recovery and complicate the transition back to the community (Finn, 1999; Travis et al., 2001). Once released from prison, an inmate who has few marketable skills and limited opportunities for employment may be more susceptible to relapse into drug and alcohol abuse and resumption of illegal activity (Laub and Sampson, 2001; Platt, 1995). A further complication is that for many inmates their physical

S. Belenko, J. Peugh / Drug and Alcohol Dependence 77 (2005) 269–281

or mental health problems make it difficult for them to sustain employment or successfully complete educational programs; 21% of state inmates report having a physical or mental health condition that limits the amount or type of work they can do (National Center on Addiction and Substance Abuse [CASA], 2002). Finally, access to affordable, drug-free housing is also important for inmates returning to the community following incarceration (Rossi, 1989; Travis et al., 2001). But many inmates face obstacles to finding adequate housing following release, due to poor family ties, lack of financial resources for a deposit on an apartment, ineligibility for public housing, or discrimination by landlords. Accordingly, additional dimensions of drug abuse and its effects need to be assessed for and considered in making clinically appropriate estimates of treatment need (McLellan et al., 1997), and in crafting appropriate treatment plans (Carise et al., 2002). Drug use has a number of other social, health, and behavioral consequences for inmates in addition to involvement in criminal behavior. Other things equal, such manifestations of drug-related consequences are likely to reflect the severity of drug use and may be related to the level of treatment need (McLellan and Alterman, 1991). For example, drug abusers who exhibit poor impulse control or anger management while under the influence may exhibit more violent behavior and thus require interventions to manage these behaviors (Dunsieth et al., 2004; Nestor, 2002; Petry, 2001). Some drug abusers may experience family disputes and employment problems as a consequence of their drug use. The ASAM Patient Placement Criteria indicate that behavioral conditions and consequences of drug use (such as educational and vocational problems, anger management problems, or motor vehicle accidents) should be taken into account in determining level of care (Mee-Lee et al., 2001). Using data from the periodic national survey of state prison inmates sponsored by the Bureau of Justice Statistics (Mumola, 1999), we analyze patterns of illegal drug use and treatment utilization among inmates and estimate the percentage who are likely to need different types of correctional drug treatment services. Informed by prior treatment matching schema, we present a framework for estimating the level of treatment need that incorporates recency and severity of drug use, drug-related behavioral consequences, and the presence of other social and health problems. The few previous efforts to estimate treatment needs among inmates (e.g. Farabee and Fredlund, 1996) typically focused solely on drug use history or psychiatric diagnoses, and we are not aware of any previous studies that have incorporated multiple dimensions to estimate the need for different types of drug treatment. The approach we take for estimating inmate needs is designed to illustrate the importance of multidimensional assessment for this population, to determine the extent to which correctional systems are meeting the treatment needs of inmates with a range of intervention options, and to identify gaps in service.

271

2. Method The data for this paper were drawn from the most recently available Survey of Inmates in State Correctional Facilities (SISCF) conducted from June through October 1997 by the U.S. Bureau of the Census for the U.S. Department of Justice, Bureau of Justice Statistics (BJS). We downloaded the disidentified raw data from this survey from a public use data set available through the National Archive of Criminal Justice Data (http://www.icpsr.umich.edu/NACJD/archive.html, Dataset #2598). BJS conducts national inmate surveys approximately every 5–7 years, and there is a lag of approximately 3 years until the data become available for public use. The survey sample was drawn from a total of 1409 State prisons listed in the 1995 Census of State and Federal Adult Correctional Facilities, or that opened prior to June 30, 1996. A stratified two-stage selection process was used that first selected prisons and then selected inmates in those prisons. In the first stage, the 13 largest male prisons and 17 largest female prisons were automatically included. The remaining 1265 male and 261 female prisons were stratified by census region and then by facility type, security level, and size of population. A systematic sample of 280 prisons (220 male and 60 female facilities) was drawn within these strata, proportionate to the size of each facility. Three facilities refused to participate and two closed before the survey was conducted, leaving a total of 275 state prisons in the final sample. For the second stage, inmates were selected using a random start and a total number of interviews based on the size and gender of the facility. A total of 14,285 interviews were conducted, with an overall response rate of 92.5%. Respondents were administered oral and written consent prior to the interview, explaining that their participation was voluntary and that all information was confidential and would not be traced back to any individual. The 1997 inmate survey used the Computer Assisted Personal Interview technique, in which the interviewer read the survey questions from a computer and entered the respondent’s answers into the computer. Because we only had access to disidentified public use data, our study was exempt from Institutional Review Board review. For the new analyses presented in this article, we applied sample weighting factors that are calculated from the probabilities that the respondent was selected for the sample, adjusting for variable nonresponse rates across selection strata, inmate respondent characteristics, and offense types. For all analyses, we applied the total sample weights to adjust for the probability that the respondent was selected for an interview, and to allow the projection of the results to the full state prison population (see Mumola (1999) for additional details on weighting procedures). These data were collected in 1997, but are the most recent representative national data on prison inmates. Although the inmate population has continued to grow since 1997, in previous work we found that there were few differences in inmate characteristics between the 1991 and 1997 surveys (CASA,

272

S. Belenko, J. Peugh / Drug and Alcohol Dependence 77 (2005) 269–281

2002). Accordingly, we believe that the 1997 data are reasonably representative of current inmates. Our analyses are different from, and expand, other analyses on substance use and treatment patterns among inmates that have used data from this survey (e.g., CASA, 2002; Mumola, 1999). The state inmate survey instrument includes questions on current and past crimes, current and past incarcerations, prison infractions, drug and alcohol use, participation in substance abuse treatment and other programs in and out of prison, and socioeconomic characteristics. Crime-related questions include details about the current offense and incarceration (e.g., when respondent was arrested, charges, time in prison), as well as offense history (previous convictions and sentences, types of prior offenses). Detailed drug use questions include lifetime use of various drugs (heroin, other opiates, cocaine, crack, amphetamines, methamphetamines, quaaludes, barbiturates, tranquilizers, phencyclidine [PCP], LSD, inhalants, marijuana), age at first use of each drug, and frequency and recency of use in the past month. Additional questions ask whether or not the offender was under the influence of the various drugs at the time he or she committed the crime that led to incarceration. Drug treatment items include previous or current involvement in drug treatment programs, including the type of program, how often and for how long the inmate had attended the program(s), and in-prison treatment participation. Alcohol use and treatment questions include frequency and quantity of alcohol use; whether the offender was under the influence of alcohol when he or she committed the crime for which they were incarcerated; and the types of prevention, education or treatment programs the offender has ever attended (or is currently attending). 2.1. Sample characteristics This sample of state inmates is 94% male, with a mean age of 33.5 years. Forty-seven percent are black non-Hispanic, 33% white non-Hispanic, 17% Hispanic, and 3.2% other race/ethnicity. Sixty percent have a high school diploma or GED, and 83% were single, divorced, separated, or widowed at the time of incarceration. Nearly half (47%) were incarcerated for a violent crime; 22% for a property crime; 21% for a drug crime; and 10% for another type. For more detailed information on inmate characteristics for the 1997 survey respondents, see CASA (2002). We also conducted the same analyses for data from the most recent national surveys of federal prison (Mumola, 1999) and local jail (Harlow, 1998) inmates, and the results were similar to those described below. For ease of presentation, we focus our analyses and discussion in this article on the findings for state inmates, which house the largest number of inmates. Details on the drug use patterns and treatment needs projections for federal and jail inmates are available from the first author. 2.2. A framework for estimating inmate treatment needs We present a “triage” model of service delivery in which comprehensive clinical assessment of drug abuse and drug-

related health and social problems are used to track inmates into different levels of treatment need. We begin with a basic axiom that the more severe the drug use, the more intensive the necessary treatment (National Institute on Drug Abuse, 1999; Thornton et al., 1998). We also assume that inmates with a greater number of other health and social problems will require more intensive intervention (McLellan and Alterman, 1991; Melnick et al., 2001). Finally, we assume that the more consequences of drug use, the greater treatment intensity that is needed. The proposed classification scheme parallels the ASAM Patient Placement Care Criteria (MeeLee et al., 2001), and other matching protocols that recognize the need for more intensive care and additional services where the drug problems and their consequences are more severe (McLellan and Alterman, 1991; Melnick et al., 2001). Other clinicians and researchers have linked a hierarchy of treatment intervention level to both severity of drug dependence and the severity of other social and health problems (McLellan and Alterman, 1991). As a conceptual framework for the treatment needs analyses we use the typical sentencing guidelines grid that is used in the federal and many state court systems (Engen and Gainey, 2000; Hofer et al., 1999; Lubitz and Ross, 2001; Marvell, 1995). In sentencing guidelines grids, two dimensions are typically used to initially determine the type and length of a sentence: the severity of the current charge, and the severity of the defendant’s prior criminal record. Although we certainly do not view increasing intensity of drug treatment as equivalent to sentencing severity (a view that would consider treatment a punishment rather than a clinical intervention to treat a disease), sentencing guidelines provide a useful framework for organizing the various dimensions that should be considered when determining level of treatment need. We start with two dimensions to determine the intensity of drug treatment needed: the severity of the inmate’s drug problem on one axis, and the number of other health and social problems on the other axis. A third dimension is imposed within each cell: whether the inmate has reported experiencing three or more drug-related consequences in their lifetime. The purpose of this other measure is to add “depth” to the estimated drug severity measure, so that we do not rely solely on quantity/frequency measures of drug use, but also take into account the extent to which the inmate has experienced negative consequences as a result of his or her drug use (no matter how minimal the drug use). This dimension can be considered as parallel to the “mitigating” or “aggravating” circumstances commonly used as a third dimension within each cell in sentencing guideline schema. Finally, we classify treatment need into four levels that correspond to the intervention programs likely to be available in prison facilities: no treatment indicated (for inmates showing no or low levels of drug use and no drug-related consequences), short-term intervention (e.g. self-help, drug education, treatment readiness, short-term motivational interventions), “outpatient” treatment (e.g. moderate-duration clinical individual and/or group counseling, but not necessarily in

S. Belenko, J. Peugh / Drug and Alcohol Dependence 77 (2005) 269–281

a separate housing unit in the correctional facility), and residential treatment (long-term intensive clinical intervention where inmates reside in a separate treatment housing unit), for inmates with recent histories of frequent hard drug use, three or more drug-related consequences, or a relatively high number of other drug-related problems). These treatment levels are roughly analogous to the broad levels of care recommended by ASAM (Mee-Lee et al., 2001). However, although ASAM patient placement criteria include many more sub-levels of treatment, correctional facilities tend to be constrained in the number of levels of care that can be offered. 2.2.1. Drug severity scale Although previous work suggests widespread drug and alcohol involvement among inmates (CASA, 2002; Mumola, 1999), it is misleading to assume that all inmates have comparable levels of problem severity or treatment need. Clinically valid diagnostic data are not available for national samples of inmates, so other measures are needed to distinguish among different levels of problem severity. The national inmate surveys do not include standardized clinical assessment tools, and therefore do not allow a clinical DSM-IV classification of drug dependence or addiction (American Psychiatric Association, 1994; Woody and Cacciola, 1997). However, the survey questions on drug use type, recency, and frequency can be used to construct a face-valid indicator of the severity of inmates’ drug involvement. Ideally, any treatment needs estimates should take into account recent drug use severity. However, although drug use certainly occurs in prisons, it is unlikely that reliable or valid data would realistically be available on current drug use for inmates (Richards and Pai, 2003); they are not likely to disclose drug use while in prison in clinical interviews or assessments. Thus, in estimating treatment needs among prison inmates or making treatment placement decisions, any drug use scale must be based on lifetime drug use and recent use prior to incarceration, reflecting drug use patterns while in the community. Accordingly, we created a scale of severity of illegal drug use. For these analyses, we focus on illegal drug use and do not incorporate alcohol use into our treatment needs projections. Nearly all inmates who have used drugs have also used alcohol, and adding alcohol use to this severity scale would not change the prevalence rates substantially. In addition, the inmate survey questions about alcohol use are different from those on drug use, making it difficult to create a severity scale that combines both (see additional discussion in ‘Study limitations’ section below). Our drug severity scale incorporates four dimensions: (1) the type of drug used (marijuana versus other illegal drugs), (2) the number of drugs used, (3) the recency of use (distinguishing use in the month prior to the offense from previous use), and (4) frequency of recent use (daily, weekly, monthly). One of the strengths of this scale is that it simultaneously takes into account several dimensions of drug use: recency, type of drug, and frequency of use.

273

The five mutually exclusive severity categories, ranked from least to most severe, are: 1. Never used “hard drugs” (defined as non-marijuana drugs including cocaine, crack, heroin, depressants, stimulants, hallucinogens, and inhalants) and did not use marijuana in the month prior to offense (could be prior marijuana user). 2. In the month prior to the offense, used marijuana but has never used hard drugs. 3. Used hard drugs, but not in the month prior to the offense. 4. In the month prior to the offense, used a single hard drug weekly or monthly. 5. In the month prior to the offense; used hard drugs daily (single or multiple) or used multiple hard drugs weekly or monthly. 2.3. Allocating Treatment Needs Table 1 shows the treatment needs allocation grid. Using our classification of drug use severity, the Y-axis is scaled from the least severe drug use category (never used hard drugs and did not use marijuana in the month prior to the offense) to the most severe category (in the month prior to the offense used hard drug[s] daily or used multiple hard drugs monthly or more often). The X-axis counts the number of other problems, ranging from 0 to 5. We further divide each cell of the grid according to whether the inmate has reported three or more drug-related consequences. Under this schema, the lower right corner of the grid represents the most severely impaired inmates, who would need the most intensive treatment. The upper left part of the grid represents the least impaired inmates, who would probably not need any treatment or fairly minimal intervention. However, this is not to say that many of these inmates may need other interventions to address educational, employment, health, or psychological problems. We made several assumptions in applying a type of treatment to each of the cells in the grid. First, we propose that any inmate in the most severe drug use category, regardless of the number of other problems or drug-related consequences, should receive residential treatment while incarcerated. Second, we assume that any inmate who has ever used nonmarijuana illegal drugs should, at a minimum, receive outpatient treatment. Third, we consider that having multiple other social or health problems implies a need for more intensive treatment than would otherwise be suggested by drug use pattern alone. Finally, we posit that having had three or more drug-related consequences generally should move an inmate up one level of treatment intensity. We recognize that these treatment level assignments are somewhat subjective and that others could make different assumptions about the types of treatment needed for inmates with different characteristics. We simply present this classification grid as one scenario to illustrate the potential treatment need in the inmate population. Correctional officials and policy makers can choose to be more or less conservative

274

S. Belenko, J. Peugh / Drug and Alcohol Dependence 77 (2005) 269–281 Table 1 Projected type of treatment needed by drug use severity, other problems, and drug-related consequences

“DC” indicates drug-related consequences. “3+ DC” indicates that the person has experienced three or more drug-related consequences prior to prison admission (see lower half of Table 3). , No tx indicated; , short-term intervention; , outpatient; , residential.

or cost-conscious in estimating the treatment needs for their correctional system. The basic underlying concept should not change, however: more extensive and recent drug use, more drug-related consequences, and a greater number of other problems should indicate a need for more intensive treatment. But, applying different assumptions may yield better or worse treatment outcomes, depending on the appropriateness of the match between level/intensity of care and treatment needs (McLellan et al., 1997).

alcohol problem screening instrument (Ewing, 1984). This percentage is similar to the 1991 BJS prison inmate survey data, which showed that 81% of state inmates were classified as substance-involved using the same definitions (Belenko and Peugh, 1999). Table 2 summarizes the percentages of inmates in each of these categories. The most common indicators of substance involvement are having ever used illegal drugs regularly (69%) or being under the influence of drugs and/or alcohol at the time of the offense (51%).

3. Results

Table 2 Categories of substance involvement among state prison inmates shown as the percentage reporting different substance-related problems prior to incarceration (n = 14,285)

3.1. Drug use, its consequences, and other health and social problems Using a broad definition, our analysis reveals that 82% of state prison inmates are involved with drugs or alcohol. They fall into one or more of the following categories: (1) they were convicted of substance-related crimes such as drug selling or driving while intoxicated, (2) were under the influence of drugs or alcohol at the time of their crime, (3) committed a crime to get money to buy drugs, or (4) had histories of regular illegal drug use (defined on the BJS inmate survey as using a drug at least weekly for a period of at least one month) or alcohol abuse. We defined alcohol abuse as having had three or more positive responses to the four-item CAGE

% regularlya

Ever used illegal drugs Convicted of a drug law violation Convicted of driving while under the influence Under the influence of drugs and/or alcohol at the time of crime Committed crime to get money to buy drugs Three or more positive CAGE responses Substance-involved inmates (percent who fit into one or more of the above categories)b

69 24 2 51 19 24 82

a Regular drug use is using a drug at least weekly for a period of at least a month. b These percentages cannot be added because of overlap.

S. Belenko, J. Peugh / Drug and Alcohol Dependence 77 (2005) 269–281 Table 3 Scale of drug use severity among state prison inmates shown as the percentage reporting level of lifetime and recent drug use (n = 14,285) % 1. Never used hard drugs and did not use marijuana in month prior to offense 2. In month prior to offense, used marijuana, but has never used hard drugs 3. Used hard drug(s) but not in the month prior to the offense 4. In month prior to offense, used a single hard drug weekly or monthly 5. In month prior to offense, used hard drug(s) daily or used multiple hard drugs weekly or monthly

30 9 25 9 27

Table 3 shows the percentages of state prison inmates in each of our drug severity scale categories. More than onequarter of state inmates (27%) fall into the most severe drug use category: they reported daily use of hard drugs, or use of multiple hard drugs one to four times, during the month prior to the offense. Overall, 36% of state inmates have recently used hard drugs and fall into the two most severe drug use categories. As the data presented in the next three sections suggest, the drug severity scale appears to have concurrent validity in that the measure correlates significantly with the number of other social and health problems, drug-related consequences, and participation in prison treatment services (see findings below). 3.1.1. Other health and social problems Other social and health problems are common and in general the prevalence of these problems is correlated with the severity of drug use. We created summary indicators of six problem areas from data available on the inmate survey: (1) psychological problems (based on the definition of Ditton (1999)—ever taken psychiatric medication, ever hospitalized for mental health problems, or ever received mental health treatment), (2) ever physically or sexually abused, (3) educational needs (lacking a high school diploma or GED at the time of arrest), (4) employment problems (unemployed at the time of arrest), and (5) housing needs (homeless or living in a group home or shelter at the time of arrest). We did not incorporate physical health problems for projecting inmate drug treatment needs. First, the inmate survey only contains a limited number of questions about recent and prior health. Second, correctional systems are constitutionally required to provide adequate health care for inmates (Estelle versus Gamble, 1976)1 , so we assume that any medical problems will be addressed regardless of drug treatment need. Among state inmates receiving a tuberculosis (TB) test since admission, 13% tested positive overall. Twenty-eight percent had been injured since admission, 7% had a health problem 1

Estelle versus Gamble, 429 U.S. 97 (1976). Available on-line: http://www.nfoweb.com/folio.pgi/ussc-1.

275

that required surgery, and 15% had at least one other medical problem (not including cold, virus, or flu). Overall, 48% of inmates had one or more medical problems; there was no significant correlation between drug use severity and number of medical problems. The top half of Table 4 displays the prevalence of mental health and social problems, indicating that the higher the severity of drug involvement, the greater the number of other problems. All correlations between type of problem or consequence are significant at p < .001. Overall, 30% of state inmates have indications of psychological problems. However, the prevalence of these problems is higher for the more severe drug use categories. Among those in the least severe drug use category (never used hard drugs and no recent marijuana use) 24% have evidence of a psychological problem. However, this prevalence increases to about one-third beginning with category 3 of the severity scale (any history of hard drug use). This pattern is similar for history of abuse. Overall, 19% of inmates report a history of sexual or physical abuse, including 57% of female inmates. Histories of such abuse are almost twice as common among those in the most severe drug use category (23%) compared to those in the least severe category (13%). Eleven percent of inmates have an indicator of a housing problem, but evidence of housing problems also rises as severity increases. While only 5% of inmates in the least severe use category had housing problems, 19% of those in the most severe drug use category have faced such problems before their arrest. In contrast, educational problems show a small negative correlation with drug use severity (39% of state inmates did not have a high school diploma or equivalent). With respect to employment problems, 31% of state inmates were unemployed at the time of incarceration. The highest prevalence of employment problems was among those in the most severe and second least severe. Combining the problem areas, more than twice as many inmates who are in the most severe drug use category have three or more of these problems (20%) than do inmates in the least severe drug use category (8%) (p < .001). 3.1.2. Drug-related consequences As expected, the prevalence of lifetime drug-related consequences increases sharply with drug use severity (lower half of Table 4).2 Each consequence, and having three or more consequence, is highly correlated with the drug severity scale (ranging from .24 to .55, all p < .001). More than half of state prison inmates ever drove a motor vehicle while under the influence of drugs, and 42% have had arguments with family or friends. Nearly one-third of these inmates have gotten into a fight while under the influence of drugs. Thirtyone percent of state inmates report at least three drug-related consequences. 2 It should be noted that the questions on the inmate survey about drugrelated consequences cover overall lifetime experiences, while the drug severity scale reflects both past and recent drug use patterns.

276

S. Belenko, J. Peugh / Drug and Alcohol Dependence 77 (2005) 269–281

Table 4 Indications of lifetime mental health and social problems, and drug-related consequences, among state prison inmates by the drug use severity scale (% within drug severity category) All inmates (n = 14,285)

Drug severity scale

%

Pearson (R)

1, least severe (n = 4111)

2 (n = 1215)

3 (n = 3533)

4 (n = 1211)

5, most severe (n = 4215)

30 19 11 40 31 13

.09a .11a .18a −.04a .10a .13a

24 13 5 43 25 8

21 11 6 49 42 11

35 22 10 34 27 13

31 21 11 40 28 13

34 23 19 39 40 20

While under the influence of drugs ever Drove a motor vehicle 53 Had a car accident 9 Had arguments with family/friends 42 Gotten into a fight 33

.51a .24a .55a .46a

12 1 5 3

65 5 37 33

62 8 45 33

70 11 56 40

80 18 76 61

Due to drugs ever Lost a job Had job or school trouble Any 3 or more of the above

.40a .42a .51a

1 1 2

4 15 24

11 20 30

19 28 40

38 47 64

Evidence of problems related to Psychological issues Sexual/physical abuse Housing problems Educational deficiencies Unemployment Any 3 or more of the above

a

15 22 31

p < .001, correlation between problem type and drug severity scale.

Almost two-thirds (64%) of state inmates in the most severe category of our drug severity scale report three or more such consequences, compared with 40% in the second-most severe category, 24% in the fourth-most severe category, and only 2% in the lowest severity category. The correlation between the severity scale and the number of drug-related consequences is .51 (p < .001). 3.2. Drug treatment utilization in state correctional systems In the absence of accurate national data on prison drug treatment participation, the extent to which prison treatment

is reaching those in need can also be measured by examining self-reported treatment participation data from the inmate survey. Table 5 shows the percentage of state inmates who reported receiving treatment since their admission, by drug severity category. Overall, only 24% of inmates report receiving any type of drug treatment since admission (including non-clinical interventions such as self-help groups or drug education programs), down from one-third of inmates in the 1991 survey (Belenko and Peugh, 1998). Overall, at the time of their interview, only 10% of state inmates report receiving any clinically- or medically-based drug treatment since admission (i.e. excluding drug education or 12-step programs).

Table 5 Self-reported treatment experience since prison admission for state inmates by type of treatment and drug severity scale (% within drug severity category) All inmates (n = 14,285)

Drug severity scale

%

1, least severe (n = 4111)

Pearson (R)

2 (n = 1215)

3 (n = 3533)

4 (n = 1211)

5, most severe (n = 4215)

Clinical/medical treatmenta Drug 10 Alcohol 8 Either type 12

.23b .12b .18b

2 4 5

7 6 9

8 9 12

13 11 15

19 13 20

Self-help/drug education Drug 20 Alcohol 21 Either type 28

.26b .10b .15b

6 14 18

17 18 24

21 23 30

30 27 36

33 24 35

7 16 20

21 21 28

25 26 35

36 32 41

40 29 42

Either clinical treatment or self-help/drug education Drug 24 .30b Alcohol 24 .12b Either type 33 .19b a b

Treatment with a clinical professional (detoxification, counseling, residential, maintenance drug). p < .001, correlation between treatment type and drug severity scale.

S. Belenko, J. Peugh / Drug and Alcohol Dependence 77 (2005) 269–281

The percentage of reported participation in any type of intervention (24%) is higher than reported in other national correctional surveys (Camp and Camp, 2004). Although relatively few inmates received clinical treatment no matter what their drug use pattern, receipt of any type of treatment is significantly correlated with the drug severity scale (p < .001). Thus, only 19% of state inmates in the most severe category had received clinical or medical drug treatment since admission, 33% self-help or drug education, and 40% either type.

277

Table 6 Projected drug treatment resource needs among state prison inmates at yearend 2002, by gender, shown as the number and percentage estimated to need different levels of treatment Male N Residential Outpatient Short-term intervention No drug treatment needed

% 353,866 210,073 226,923 332,521

Total

Female

1,123,383

31.5 18.7 20.2 29.6 100

N

%

45,112 13,974 7,159 20,012 86,257

52.3 16.2 8.3 23.2 100

3.3. Projected state correctional treatment needs The preceding analyses clearly point to high rates of prior drug involvement among incarcerated populations, linked to a high prevalence of drug-related behavioral consequences, and other social and health problems. At the same time, partitioning the drug-involved inmate population on the basis of a face-valid severity/recency scale suggests that only about one-quarter of state inmates clearly need intensive treatment, and that an additional 30–40% might benefit from less intensive treatment. It is also apparent that relatively few inmates with drug abuse problems receive clinical interventions while incarcerated. The inmate population is heterogeneous: inmates have different intensities of drug involvement, and different constellations of other problems that may require service intervention. We calculated prevalence estimates for each of the cells in the treatment needs grid for state correctional systems. Prevalence rates were separately estimated for male and female inmates, because most states house males and females in separate facilities and treatment programs would have to be sited separately as well. In order to estimate the actual number of treatment slots needed, we used the state inmate populations as of December 2002 (Harrison and Beck, 2003), and applied the prevalence estimates calculated for each of

the cells in the Table 1 grid to these figures (Table 6). There were 1,123,383 male and 86,257 female state inmates at the end of 2002. Our analyses yield an estimate that at the end of 2002, approximately 353,866 male inmates (31.5%) needed residential treatment and 210,073 (18.7%) needed outpatient treatment. In addition, an estimated 45,112 females (52.3%) needed residential and 13,974 (16.2%) needed outpatient treatment. Female state prison inmates have a much higher estimated need for residential treatment than males, primarily because of a higher prevalence of other problems (Peugh and Belenko, 1999). We estimate that 29.6% of male state inmates and 23.2% of females need no drug treatment intervention. These treatment needs projections do not necessarily translate directly into an equivalent number of available treatment beds or slots. Estimating the true gap between current treatment availability and treatment need requires a consideration of the length and timing of treatment. For example, correctional residential treatment programs typically last for 6–12 months, and many inmates participate in these programs in their last year of incarceration. Accordingly, although we estimate that 353,866 male prison inmates in 2002 needed residential treatment, fewer residential beds would actually

Table 7 Treatment experience since admission for state prison inmates, by level of treatment, gender, and projected type of treatment intervention needed (% within projected treatment need category) Male No treatment indicated (n = 3367) Clinical/medical treatment Drug 1 Alcohol 4 Either type 5 Self-help/drug education Drug 6 Alcohol 14 Either type 18

Female Short-term intervention (n = 2300)

Outpatient (n = 2115)

Residential (n = 3562)

6 6 9

10 9 13

18 13 19

1 3 3

17 20 26

25 25 32

32 25 35

29 28 37

39 30 42

Either clinical treatment or self-help/drug education Drug 7 20 Alcohol 16 23 Either type 20 30

No treatment indicated (n = 687)

Short-term intervention (n = 244)

Outpatient (n = 492)

Residential (n = 1518)

7 5 9

11 8 15

23 13 23

4 9 13

16 15 22

24 18 30

34 20 35

5 11 15

20 16 26

28 21 35

43 25 44

278

S. Belenko, J. Peugh / Drug and Alcohol Dependence 77 (2005) 269–281

be needed because of turnover and the varying points in their incarceration period when these inmates would actually enter residential treatment. Detailed treatment bed needs analyses would be required to estimate the number of additional individual treatment slots needed by state correctional systems. As a test of the validity of our treatment needs framework, we compared inmates’ projected treatment need to the treatment actually received since admission. Table 7 shows that our treatment placement scheme is highly correlated with the likelihood of having received some type of substance abuse intervention while incarcerated. It is not clear whether this reflects active assessment and placement by correctional facilities, or that inmates with more extensive drug involvement and related problems are more likely to seek out services on their own. However, an extensive treatment gap is still suggested by these data: among state inmates projected to need residential treatment, only19% of males and 23% of females received any clinical intervention since admission, and about one-third were involved in some type of self-help group or drug education.

4. Discussion This paper has described drug use patterns among state prison inmates, and presented a protocol for estimating the levels of their drug treatment needs, drawing on the ASAM Patient Placement Criteria and other matching schema as guides for incorporating different dimensions of drug use severity and its consequences. With a conceptual framework drawn from the sentencing guidelines used in many criminal court systems, our matching criteria use three different dimensions of illegal drug use and related problems: the severity of drug use based on the type/frequency/recency of use, the number of other social and health problems, and the number of drug-related behavioral consequences. Our findings suggest that inmates need a range of treatment modalities, and that the existing delivery of correctional treatment, especially residential, is highly inadequate relative to need. We found high levels of drug involvement among state inmates, but considerable variation in the severity and recency of such use, as well as its health and social consequences. We estimate that about one-third of male and more than half of female state prison inmates need long-term residential treatment. Although inmates in our most severe drug use categories are more likely to have received treatment while incarcerated, only about one-fifth received any clinical treatment services. On the other hand, about half of male and one-third of female inmates may need no treatment or only a short-term intervention. By expanding the currently limited range of treatment levels and modalities offered to inmates, and conducting more comprehensive clinical assessments, correctional systems can meet their treatment needs in a more cost-effective manner. Given the declining resources available for correctional treatment (Belenko and Peugh, 1998), and the increasing role of behavioral managed care models in

delivering treatment and other services in inmates (Godbole et al., 1998; Morrissey, 1996; Packer, 1998; Patterson, 1998), matching need to treatment level is a key strategy for correctional systems to consider. The higher levels of projected treatment need among women inmates are consistent with other studies that find multiple health and social problems among female offenders (Belenko and Peugh, 1999; Mahan, 1996; McClellan et al., 1997; Prendergast et al., 1995; Wellisch et al., 1993). Accordingly, there is an urgent need to increase the attention to gender-specific treatment needs of female inmates, and to expand treatment capacity in women’s correctional facilities. In addition to the need to expand multi-modality correctional treatment, there are other implications from our findings that may present challenges for correctional treatment providers, policy makers, and researchers. Correctional systems need to implement expanded and improved multidomain assessment instruments, not only upon admission, but within a year of release. Formal, standardized treatment placement criteria should be developed, implemented, and evaluated in state correctional systems. Research on the impacts of correctional treatment needs to expand to include nonresidential programs, and long-term follow-up in the community with multiple outcome measures (Harrison, 2001). In addition to improved assessment, more effective inmate treatment placement decisions could be facilitated by expanding clinical evaluation of their readiness to change (De Leon, 1996; Prochaska and DiClemente, 1986) and treatment motivation (Joe et al., 1998; Shen et al., 2000; Simpson and Joe, 1993). Finally, correctional treatment should focus on the identified health or social problems in addition to drug abuse or dependence, and should incorporate transitional planning for continued care following release from prison or jail. 4.1. Study limitations The data for these analyses were derived from self-reports of drug use, drug-related problems, other social and health problems, and prison treatment experience. As with any self-reported data, the validity of the subjects’ responses may be affected by underreporting or overreporting of socially undesirable behaviors, difficulties recalling past behaviors, misunderstanding of the survey questions, or other errors. The survey responses were not validated by drug tests, clinical assessment tools, interviews with counselors or family members, or institutional records. Inmates may overestimate treatment participation by including other mental health counseling or health education received while in prison, or may underreport participation because of recall problems. Inmates may also utilize in-prison treatment for reasons other than help with a drug problem-to reside in a more stable housing unit, to demonstrate good conduct, or to be eligible for early release, for example. Other limitations of the national inmate survey data should be mentioned. First, the available indicators of drug sever-

S. Belenko, J. Peugh / Drug and Alcohol Dependence 77 (2005) 269–281

ity, drug-related consequences, and other health and social problems were based on a combination of lifetime and recent behaviors. Accordingly, the survey data do not allow an analysis of the temporal relationships among these factors. More extensive “life history” interviews would be needed to clarify the causal and temporal relationships among drug use patterns and the consequences and problems associated with such use. Second, the most recent inmate surveys were conducted about 7 years ago and may not reflect current inmate drug use and other problem profiles. However, in our analyses of previous inmate surveys, we found that patterns of drug use and other social and health problems were relatively stable between the 1989 and 1991, and 1995 and 1997 jail and prison inmate surveys (CASA, 2002). As discussed earlier, we focused on drug treatment and did not take into account alcohol use. We assume that drug treatment interventions will also deal with alcohol problems, and separate analyses would be needed to estimate the number of inmates needing different types of alcohol treatment. The need to treat inmate alcohol abuse as an independent problem, or as a mediating factor in drug treatment outcomes, should be considered in subsequent research and clinical studies on this issue. Indications of alcohol problems are more likely among inmates in the more severe drug use categories: among state prison inmates 34% of those in our most severe drug category and 31% of those in the second most severe category have three or more positive CAGE responses, indicating a high likelihood of alcohol abuse or dependence (Allen and Columbus, 1995). In contrast, only 13% of inmates in the two least severe categories have three or more positive CAGE responses. Finally, motivation to change and recovery status are potentially important factors in making treatment placement decisions (De Leon, 1996; Prochaska and DiClemente, 1986). However, the inmate survey did not include data on recovery status, readiness to change, or treatment motivation. Such factors could be important in treatment outcomes and successful treatment matching, and future research should incorporate these measures in estimating treatment needs and developing placement criteria for inmates. 4.2. Conclusions States have responded to the dearth of correctional treatment capacity by expanding community-based treatment alternatives such as drug courts (Belenko, 2002; Marlowe, 2003) or Proposition 36-type diversion programs (Hser et al., 2003; Jett, 2001). However, there are several indications that such treatment is not likely to address the substantial treatment gap for prison inmates. First, although there are more than 1000 operational drug courts, they only serve about 5% of the potentially eligible offender population, and tend to target nonviolent offenders who are unlikely to be sentenced to state prison (Belenko, 2002). Second, the presence of mandatory sentencing laws for drug offenders means that the growth in the drug-involved prison population is unlikely to abate in

279

the foreseeable future (Blumstein and Beck, 1999). Finally, relatively few states have adopted Proposition 36-style programs, and the early evidence is mixed on their ability to increase offender engagement in long-term treatment (Hser et al., 2003). Given the large numbers of drug-involved inmates, the limited treatment capacity in prisons, and the continuing reliance on incarceration in the United States, the substantial gap in services relative to need indicated by our data is likely to continue into the foreseeable future. Coupled with the range of drug use severity and associated problems that we found among prison inmates, improving treatment matching in this population should be an important policy goal. We recognize that effective patient placement has been difficult to achieve in community settings (Gastfriend and McLellan, 1997; Project MATCH, 1998), and that the challenges may be even greater in correctional systems. In addition to improving clinical assessment and expanding it to incorporate drug-related health and social problems (Knight et al., 2002), and substantially expanding the availability of different levels of care (as well as continuing care following release), incentives and other contingencies for inmate treatment engagement may be needed. Providing early release from prison, vouchers, inmate housing incentives, continuing care linkages, or step-down to lower security levels, are strategies that can be tested for increasing the use of available treatment. Although the initial funding outlay and logistical issues would be considerable, increasing access to different levels of treatment could provide substantial long-term economic and social benefits from reduced recidivism, easier transition to the community following release, and reduced drug abuse (Belenko and Peugh, 1998; Carey and Finigan, 2004; Jofre-Bonet and Sindelar, 2001; Knight et al., 1999; Martin et al., 1999).

Acknowledgements The research for this paper was supported in part by grant #2000-IJ-CX-0019 from the National Institute of Justice to the National Center on Addiction and Substance Abuse at Columbia University (Steven Belenko, principal investigator). The research was conducted when the first author was affiliated with the National Center on Addiction and Substance Abuse at Columbia University, and the second author was affiliated with the New York State Office of Alcohol and Substance Abuse Services. The points of view and conclusions presented in this paper are those of the authors and do not necessarily represent the views of the National Institute of Justice, the Treatment Research Institute at the University of Pennsylvania, the National Center on Addiction and Substance Abuse at Columbia University, or Harris Interactive. We appreciate the helpful comments and suggestions made by Dr. A. Thomas McLellan, and by three anonymous reviewers.

280

S. Belenko, J. Peugh / Drug and Alcohol Dependence 77 (2005) 269–281

References Allen, J., Columbus, M. (Eds.), 1995. Assessing Alcohol Problems: A Guide for Clinicians and Researchers. National Institute on Alcohol Abuse and Alcoholism, Bethesda, MD. American Psychiatric Association, 1994. Diagnostic and Statistical Manual of Mental Disorders, fourth ed. American Psychiatric Association Press, Washington, DC (DSM-IV). Anno, J., 1991. Prison Health Care: Guidelines for the Management of an Adequate Delivery System. U.S. Department of Justice, National Institute of Corrections, National Commission on Correctional Health Care, Washington, DC. Beck, A.J., Maruschak, L., 2001. Mental Health Treatment in State Prisons, 2000. Pub. No. NCJ 188215. U.S. Department of Justice, Bureau of Justice Statistics, Washington, DC. Belenko, S., 2002. Drug courts. In: Leukefeld, C., Tims, F., Farabee, D. (Eds.), Treatment of Drug Offenders: Policies and Issues. Springer, New York, pp. 301–318. Belenko, S., Peugh, J., 1998. Fighting crime by treating substance abuse. Issues Sci. Technol. 15, 53–60. Belenko, S., Peugh, J., 1999. Behind Bars: Substance Abuse and America’s Prison Population: Technical Report. National Center on Addiction and Substance Abuse at Columbia University, New York. Belenko, S., Lang, M.A., O’Connor, L.A., 2003. Self-reported psychiatric treatment needs among felony drug offenders. J. Contemp. Crim. Just. 19, 9–29. Blumstein, A., Beck, A.J., 1999. Population growth in U.S. prisons, 1980–1996. In: Tonry, M., Petersilia, J. (Eds.), Prisons, vol. 26 of Crime and Justice: A Review of Research, University of Chicago Press, Chicago. Camp, G., Camp, C., 2004. The Corrections Yearbook, 2002. Criminal Justice Institute, Middletown, CT. Carey, S.M., Finigan, M.W., 2004. A detailed cost analysis in a mature drug court setting. J. Contemp. Crim. Just. 20, 315–338. Carise, D., Gurel, O., Kendig, C., McLellan, A.T., 2002. Giving Clinical Meaning to Patient Assessment: Technology Transfer to Improve Treatment Care Planning and Service Delivery. Treatment Research Institute, Philadelphia, PA. Deitch, D., Carleton, S., Koutsenok, I., Marsolais, K., 2002. Therapeutic community treatment in prisons. In: Leukefeld, C., Tims, F., Farabee, D. (Eds.), Treatment of Drug Offenders: Policies and Issues. Springer, New York, pp. 127–137. De Leon, G., 1996. Integrative recovery: a stage paradigm. Subst. Abuse 17, 51–63. Ditton, P.M., 1999. Mental Health and Treatment of Inmates and Probationers: Special Report. Pub. No. NCJ 174463. U.S. Department of Justice, Bureau of Justice Statistics, Washington, DC. Dunsieth Jr., N.W., Nelson, E.B., Brusman-Lovins, L., Holcomb, J.L., Beckman, D., Welge, J.A., Roby, D., Taylor Jr., P., Soutullo, C.A., McElroy, S.L., 2004. Psychiatric and legal features of 113 men convicted of sexual offenses. J. Clin. Psychiatry 65, 293– 300. Engen, R.L., Gainey, R.R., 2000. Modeling the effects of legally relevant and extralegal factors under sentencing guidelines: the rules have changed. Criminology 38, 1207–1229. Etheridge, R.M., Craddock, S.G., Dunteman, G.H., Hubbard, R.L., 1995. Treatment services in two national studies of community-based drug abuse treatment programs. J. Subst. Abuse 7, 9–26. Ewing, J., 1984. Detecting alcoholism: the CAGE questionnaire. J. Am. Med. Assoc. 252, 1906–1907. Farabee, D., Fredlund, E., 1996. Self-reported drug use among recently admitted jail inmates: estimating prevalence and treatment needs. Subst. Use Misuse 31, 423–435. Finn, P., 1999. Washington State’s Corrections Clearinghouse: A Comprehensive Approach to Offender Employment. National Institute of Justice Program Focus, U.S. Department of Justice, National Institute of Justice, Washington, DC.

Gastfriend, D.R., McLellan, A.T., 1997. Treatment matching: theoretical basis and practical implications. Med. Clin. North Am. 81, 945–966. Godbole, A., Temkin, T., Cradock, C., 1998. Restructuring public mental health and substance abuse service systems. J. Health Care Fin. 24, 16–26. Hammett, T.M., Gaiter, J.L., Crawford, C., 1998. Reaching seriously at-risk populations: health interventions in criminal justice settings. Health Educ. Behav. 25, 99–120. Hammett, T., Harmon, P., Maruschak, L., 1999. 1996–1997 Update: HIV, AIDS, STDs, and TB in Correctional Facilities. U.S. Department of Justice, National Institute of Justice, and Centers for Disease Control and Prevention, Washington, DC. Harlow, C.W., 1998. Profile of Jail Inmates, 1996. Pub. No. NCJ 164620. U.S. Department of Justice, Bureau of Justice Statistics, Washington, DC. Harrison, L., 2001. The revolving prison door for drug-involved offenders: challenges and opportunities. Crime Delinq. 47, 462–484. Harrison, P.M., Beck, A.J., 2003. Prisoners in 2002. Pub. No. NCJ 200248. U.S. Department of Justice, Bureau of Justice Statistics, Washington, DC. Hofer, P.J., Blackwell, K.R., Ruback, R.B., 1999. The effect of federal sentencing guidelines on inter-judge sentencing disparity. J. Crim. Law Criminol. 90, 239–321. Hoffmann, N., Halikas, J., Mee-Lee, D., et al., 1991. American Society of Addiction Medicine—Patient Placement Criteria for the Treatment of Psychoactive Substance Use Disorders. American Society of Addiction Medicine, Washington, DC. Hser, Y.-I., Polinsky, M.L., Maglione, M., Anglin, M.D., 1999. Matching clients’ needs with drug treatment services. J. Subst. Abuse Treat. 16, 299–305. Hser, Y.-I., Teruya, C., Evans, E., Longshore, D., Grella, C., Farabee, D., 2003. Treating drug-abusing offenders: initial findings from a fivecounty study on the impact of California’s Proposition 36 on the treatment system and patient outcomes. Eval. Rev. 27, 479–505. Institute of Medicine, 1990a. Broadening the Base of Treatment for Alcohol Problems. National Academy Press, Washington, DC. Institute of Medicine, 1990b. Treating Drug Problems. National Academy Press, Washington, DC. Jett, K.P., 2001. Proposition 36: issues and challenges. J. Psychoactive Drugs 33, 321–328. Joe, G., Simpson, D., Broome, K., 1998. Effects of readiness for drug abuse treatment on client retention and assessment of process. Addiction 93, 1177–1190. Jofre-Bonet, M., Sindelar, J.L., 2001. Drug treatment as a crime fighting tool. J. Mental Health Policy Econ. 4, 175–188. Knight, K., Simpson, D.D., Hiller, M.L., 1999. Three-year reincarceration outcomes for in-prison therapeutic community treatment in Texas. Prison J. 79, 337–351. Knight, K., Simpson, D.D., Hiller, M.L., 2002. Screening and referral for substance-abuse treatment in the criminal justice system. In: Leukefeld, C., Tims, F., Farabee, D. (Eds.), Treatment of Drug Offenders: Policies and Issues. Springer, New York, pp. 259–272. Lamb, H.R., Weinberger, L.E., 1998. Persons with severe mental illness in jails and prisons: a review. Psychiatric Serv. 49, 483–492. Laub, J.H., Sampson, R.J., 2001. Understanding desistance from crime. In: Tonry, M. (Ed.), Crime and Justice. University of Chicago Press, Chicago, pp. 1–69. Lubitz, R., Ross, T., 2001. Sentencing Guidelines: Reflections on the Future. U.S. Department of Justice, National Institute of Justice, Washington, DC. Mahan, S., 1996. Crack Cocaine, Crime, and Women. Sage Publications, Thousand Oaks, CA. Marlowe, D.B., 2003. Integrating substance abuse treatment and criminal justice supervision. NIDA Sci. Pract. Perspect. 2, 4–14. Marquart, J., Merianos, D., Hebert, J., Carroll, L., 1997. Health conditions and prisoners: a review of research and emerging areas of inquiry. Prison J. 77, 184–208.

S. Belenko, J. Peugh / Drug and Alcohol Dependence 77 (2005) 269–281 Martin, S.S., Butzin, C.A., Saum, C.A., Inciardi, J.A., 1999. Three-year outcomes of therapeutic community treatment for drug-involved offenders in Delaware: from prison to work release to aftercare. Prison J. 79, 294–320. Marvell, T.B., 1995. Sentencing guidelines and prison population growth. J. Crim. Law Criminol. 85, 696–709. Mattson, M.E., Allen, J.P., Longabaugh, R., Nickless, C.J., Connors, G.J., Kadden, R.M., 1994. A chronological review of empirical studies matching alcoholic clients to treatment. J. Stud. Alc. 12 (Suppl.), 16–29. McClellan, D.S., Farabee, D., Crouch, B.M., 1997. Early victimization, drug use, and criminality: a comparison of male and female prisoners. Crim. Just. Behav. 24, 455–476. McLellan, A.T., Luborsky, L., Woody, G.E., O’Brien, C.P., Druley, K.A., 1983. Predicting response to alcohol and drug abuse treatment. Arch. Gen. Psychiatry 40, 620–625. McLellan, A.T., Alterman, A.I., 1991. Patient treatment matching: a conceptual and methodological review with suggestions for future research. In: NIDA Research Monograph no. 106. U.S. Department of Health and Human Services, National Institute on Drug Abuse, Rockville, MD, pp. 114–135. McLellan, A.T., Arndt, I.O., Metzger, D.S., O’Brien, C.P., 1993. The effects of psychosocial services in substance abuse treatment. J. Am. Med. Assoc. 269, 1953–1959. McLellan, A.T., Grissom, G.R., Zanis, D., Randall, M., Brill, P., O’Brien, C.P., 1997. Problem service ‘matching’ in addition treatment. Arch. Gen. Psychiatry 54, 730–735. Mee-Lee, D., Shulman, G., Fishman, M., Gastfriend, D.R., Griffith, J.H. (Eds.), 2001. ASAM patient placement criteria for the treatment of substance-related disorders, Second ed.-revised (ASAM PPC-2R). American Society of Addiction Medicine, Chevy Chase, MD. Melnick, G., De Leon, G., Thomas, G., Kressel, D., 2001. A clienttreatment matching protocol for therapeutic communities: first report. J. Subst. Abuse Treat. 21, 119–128. Moos, R.H., Finney, J.W., 1995. Substance abuse treatment programs and processes: linkages to patients’ needs and outcomes. J. Subst. Abuse 7, 1–8. Morrissey, J.P., 1996. Shifting Boundaries, Reducing Costs: Managed Behavioral Health Care and the Criminal Justice System. Prepared for: The National GAINS Center for People with Co-Occurring Disorders in the Justice System, Policy Workshop Retreat, St. Michael’s Maryland, January 23–24, 1996. Mumola, C.J., 1999. Substance Abuse and Treatment of State and Federal Prisoners, 1997. Pub. No. NCJ 172871. U.S. Department of Justice, Bureau of Justice Statistics, Washington, DC. National Center on Addiction and Substance Abuse at Columbia University (CASA), 2002. Trends in Substance Abuse and Treatment Needs among Inmates. Final report to the National Institute of Justice, The National Center on Addiction and Substance Abuse at Columbia University, New York. National Institute on Drug Abuse, 1999. Principles of Drug Addiction Treatment: A Research-Based Guide. U.S. Department of Health and Human Services, National Institute on Drug Abuse, Bethesda, MD. Nestor, P.G., 2002. Mental disorder and violence: personality dimensions and clinical features. Am. J. Psychiatry 159, 1973–1978. Packer, I.K., 1998. Privatised managed care and forensic mental health services. J. Am. Acad. Psychiatry Law 26, 123–129. Patterson, R.F., 1998. Managed behavioral healthcare in correctional settings. J. Am. Acad. Psychiatry Law 26, 467–473.

281

Petersilia, J., 1999. Parole and prisoner reentry in the United States. In: Wilson, J.Q., Tonry, M. (Eds.), Crime and Justice, vol. 26. University of Chicago Press, Chicago. Petry, N.M., 2001. Substance abuse, pathological gambling, and impulsiveness. Drug Alcohol Depend. 63, 29–38. Peugh, J., Belenko, S., 1999. Substance-involved women inmates: challenges to providing effective treatment. Prison J. 79, 23–44. Platt, J.L., 1995. Vocational rehabilitation of drug abusers. Psych. Rev. 117, 416–433. Prendergast, M.L., Wellisch, J., Falkin, G.P., 1995. Assessment of and services for substance-abusing women offenders in community and correctional settings. Prison J. 75, 240–256. Prochaska, J., DiClemente, C., 1986. Toward a comprehensive model of change. In: Miller, W.R., Heather, N. (Eds.), Treating Addictive Behaviors: Processes of Change. Plenum Press, New York. Project MATCH Research Group, 1998. Matching alcoholism treatment to client heterogeneity: Project MATCH three-year drinking outcomes. Alcohol. Clin. Exp. Res. 22, 1300–1311. Richards, H.J., Pai, S.M., 2003. Deception in prison assessment of substance abuse. J. Subst. Abuse Treat. 24, 121–128. Rossi, P.H., 1989. Down and Out in America: Origins of Homelessness. University of Chicago Press, Chicago. Rychtarik, R., Connors, G.J., Whitney, R.B., McGillicuddy, N.B., Fitterling, J.M., Wirtz, P.W., 2000. Treatment settings for persons with alcoholism: evidence for matching clients to inpatient versus outpatient care. J. Consult. Clin. Psych. 68, 277–289. Shen, Q., McLellan, A.T., Merrill, J.C., 2000. Clients’ perceived need for treatment and its impact on outcome. Subst. Abuse 21, 179– 192. Simpson, D., Joe, G., 1993. Motivation as a predictor of early dropout from drug abuse treatment. Psychotherapy 30, 357–368. Simpson, D.D., Joe, G.W., Fletcher, B.W., Hubbard, R.L., Anglin, M.D., 1999. A national evaluation of treatment outcomes for cocaine dependence. Arch. Gen. Psychiatry 56, 510–514. Taxman, F., 1999. Unraveling what works for offenders in substance abuse treatment services. Natl. Drug Court Inst. Rev. 2, 91–132. Teplin, L., 1994. Psychiatric and substance abuse disorders among male urban jail detainees. Am. J. Pub. Health 84, 290–293. Teplin, L.A., Abram, K.M., McClelland, G.M., 1996. Prevalence of psychiatric disorders among incarcerated women. Arch. Gen. Psychiatry 53, 505–512. Thornton, C.C., Gottheil, E., Weinstein, S.P., Kerachsky, R.S., 1998. Patient-treatment matching in substance abuse: drug addiction severity. J. Subst. Abuse Treat. 15, 505–511. Turner, W.M., Turner, K.H., Reif, S., Gutowski, W.E., Gastfriend, D., 1999. Feasibility of multidimensional substance abuse treatment matching: automating the ASAM patient placement criteria. Drug Alcohol Depend. 55, 35–43. Travis, J., Solomon, A.L., Waul, M., 2001. From Prison to Home: The Dimensions and Consequences of Prisoner Reentry. The Urban Institute, Washington, DC. Wellisch, J., Anglin, M.D., Prendergast, M.L., 1993. Treatment strategies for drug-abusing women offenders. In: Inciardi, J.A. (Ed.), Sage Criminal Justice System Annuals: Drug Treatment and Criminal Justice, vol. 27. Sage Publications, Newbury Park, CA, pp. 5–25. Woody, G.E., Cacciola, J., 1997. Diagnosis and classification: DSM-IV and ICD-10. In: Lowinson, J.H., Ruiz, P., Millman, R.B., Langrod, J.G. (Eds.), Substance Abuse: A Comprehensive Textbook. Williams & Wilkins, Baltimore, pp. 361–363.