Addiction Severity Index Scores and Urine Drug Screens at Baseline as Predictors of Graduation From Drug Court

Abstract

Graduation rates from drug courts are impressive and are often attributed to the delivered treatments. However, it is unclear whether graduation rates are bolstered by low severity of drug use problems upon entry into drug court. To address this question, this study examined the relationship between baseline substance use severity and graduation rates among 251 drug court clients. Results revealed that participants with subthreshold drug composite scores on the Addiction Severity Index (ASI) were significantly more likely to graduate than those scoring in the mild-to-moderate or severe range. Furthermore, results revealed that participants who provided a drug-negative baseline urine were significantly more likely to graduate than those who provided a drug-positive baseline urine. A binary logistic regression analysis revealed that ASI drug composite score, urine screen, race, and years educated were statistically significant predictors of drug court graduation.

Keywords

drug court Addiction Severity Index urine drug screen substance use criminal justice

The co-occurrence of drug use and crime presents unique challenges for offenders, criminal justice professionals, and treatment providers. Drug courts are a relatively novel strategy that combines elements of punitive- and treatment-oriented approaches. They are separate criminal court dockets that provide judicially supervised treatment and case management services in lieu of criminal prosecution or incarceration. Drug courts are a type of “problem-solving court” rooted in the principles of therapeutic jurisprudence, and they provide wide-ranging treatment and case management services to drug-involved offenders under the watchful eye of the drug court judge. Drug courts typically run between 6 months and 1 year, and drug court clients participate in a combination of treatment, relapse prevention, monitoring, and enforcement (Walsh, 2011). In addition, behavioral modification in the form of rewards and sanctions are commonly used in a treatment team approach. (Peyton & Gossweiler, 2001).

Over the past 20 years, drug courts’ success in reducing drug use and criminal recidivism has been thus far supported by controlled research, program evaluations, and meta-analyses (see, for example, Belenko, 1998, 1999, 2001, 2002; Belenko, DeMatteo, & Patapis, 2007; Government Accountability Office, 2011; Mitchell, Wilson, Eggers, & MacKenzie, 2012; Shaffer, 2011; Wilson, Mitchell, & Mackenzie, 2006). Over the past decade, researchers have examined the “drug court black box” (Goldkamp, White, & Robinson, 2001) meaning that the focus of research has increasingly been on “how” drug courts are effective. Some recent studies have examined the contributions of the individual operative components of drug courts to successful outcomes. One meta-analysis by Shaffer (2011) found that target population, program leverage and intensity, and staff characteristics account for most of the variability in drug court effectiveness. Another recent meta-analysis by Mitchell et al. (2012) found that only program leverage, intensity, and severity of offender charges slightly moderated effectiveness.

Nationally, drug court graduation rates average an impressive 50% to 70% (General Accounting Office, 1997). While this is a promising statistic, understanding the differences between those who are able to meet the requirements and those who fail to graduate could increase the efficiency of drug courts. In addition, clients who do not complete a drug court typically return to standard criminal justice processing where costs increase as they face prosecution, sentencing, or a parole violation (Belenko, Patapis, & French, 2005; Bhati, Roman, & Chalfin, 2008). The effectiveness and efficiency of drug courts could be enhanced by identifying—early and reliably—clients who are likely to succeed or fail, as greater reductions in recidivism are found in drug courts with higher graduation rates (Mitchell et al., 2012). Of particular interest to the present study is whether severity of substance use at baseline could to some degree account for drug court effectiveness as measured by successful completion of drug court. If so, this may provide some credence to the growing body of literature suggesting that dosages or intensity of treatment could be modified for different types of offenders.

Severity of Substance Use Problems Upon Entry

Although the National Association of Drug Court Professionals (NADCP; 1997) guidelines suggest matching clients to treatments based on specific needs, drug courts often provide the same services to all clients. For example, a national survey found that the minimum expected length of participation does not vary for most drug courts (70.5%) despite differences among participants in drug use severity (Rossman, Roman, Zweig, Rempel, & Lindquist, 2011). In effect, some drug court clients may receive a higher intensity and dosage of treatment than needed. For instance, some participants are nonaddicted adolescents or young adults just beginning to experiment with drugs and others are drug dealers with no drug use history (Marlowe, Patapis, & DeMatteo, 2003). Some studies have shown that almost one half of misdemeanor clients (Marlowe, Festinger, & Lee, 2003) and one third of felony clients (Marlowe, Festinger, & Lee, 2004) received subthreshold drug composite scores on the Addiction Severity Index (ASI). As a benchmark, those scores do not significantly differ from community samples of individuals not seeking treatment. Other studies measuring predictors of drug court graduation have found less severity of drug use before arrest to significantly predict graduation (Jones & Kemp, 2011; Sechrest & Shicor, 2001). Furthermore, data suggest that positive outcomes are related to drug courts’ capacity to adapt treatment (Koob, Brocato, & Kleinpeter, 2011). As such, there have been calls for increased tailoring of programs (e.g., Jones & Kemp, 2011; Marlowe et al., 2008). As a whole, one may question whether some drug court graduates succeed as a result of their subthreshold clinical needs or the treatment/services provided.

Two measurements that may aid drug court personnel in identifying baseline substance use severity are the ASI and drug urine screens. The ASI assesses severity of problems (lifetime and past 30 days) in seven areas: medical, employment/support, alcohol use, drug use, illegal activity, family relations, and psychiatric condition. Higher composite scores suggest a more severe problem that requires more intensive treatment. Some research suggests the ASI is not a useful predictor of treatment dropout, specifically in samples of cocaine-dependent individuals (Alterman, McKay, Mulvaney, & McLellan, 1996; Alterman et al., 1997). Nevertheless, the ASI is widely used in drug courts with individuals who use an array of substances and is therefore important to further understand whether there is a relationship between ASI scores and drug court success. Those with low ASI drug abuse composite scores may represent individuals just beginning to experiment with drugs or nonusing drug dealers (Marlowe, Patapis, et al., 2003).

Unlike ASI scores, there is evidence that baseline urine screens are robust predictors of attrition from outpatient treatment programs. Urine specimens testing positive for cocaine at baseline predicted attrition from outpatient treatment (Alterman et al., 1996; Higgins et al., 1993). Alterman et al. (1997) found that patients in 6-month outpatient treatment who had a baseline drug-positive urine specimen for cocaine were more likely to have unfavorable drug treatment outcomes than those who did not test positive for cocaine. As for predicting early termination, Ehrman, Robbins, and Cornish (2001) found that a single drug-positive urine specimen for cocaine prior to a 4-week treatment predicted the likelihood and frequency of cocaine use over the subsequent 1-month period.

Although researchers have examined the relationship between urine drug screens and drug court performance/graduation, the relationship between baseline urine screens and drug court completion has not been examined. This is surprising given the ability of one urine screen to predict treatment success in other contexts. The drug court research suggests that clients who consistently produce negative drug screens are more likely to graduate. For instance, one study found that “optimal performers” in drug courts were those who consistently had negative drug screens and were more likely to graduate and remain drug-free 6 months after graduation (DeMatteo, Marlowe, Festinger, & Arabia, 2009). More recently, a study found that the likelihood of graduation from drug court was significantly lower for participants with the poorest early-phase drug use patterns (Jones & Kemp, 2011). These findings serve as a basis for further inquiries into the implications of a baseline drug-positive urine specimen on treatment success.

Other Factors Associated With Drug Court Completion

Baseline substance use severity may be an indicator of drug court success, but several factors have been studied in the context of drug court graduation rates. Some studies have looked at certain demographic variables in predicting drug court success. We are not suggesting that these characteristics should be used to identify offenders eligible for drug courts because it “would almost certainly run afoul of due process and equal protection requirements to exclude individuals from correctional rehabilitation programs based upon their immutable demographic characteristics” (Marlowe, Patapis, et al., 2003, p. 43). However, identifying characteristics of drug court clients likely to succeed could help drug courts tailor treatment to individual needs.

Age, Gender, and Race

Research suggests older drug court clients graduate at higher rates than younger clients (Hickert, Boyle, & Tollefson, 2009; Saum, Scarpitti, & Robbins, 2001). Findings regarding gender are less conclusive, as few studies have examined the efficacy of drug courts for female offenders. Most studies have found no significant gender differences in relation to graduation (Peters, Haas, & Murrin, 1999; Saum et al., 2001; Schiff & Terry, 1997; Sechrest & Shicor, 2001); however, the General Accounting Office (1997) found that males are more successful in completing drug courts. Butzin, Saum, and Scarpitti (2002) suggested that female offenders encounter more barriers to success, including employment and income problems and a higher likelihood of mental health issues (e.g., depression, anxiety, suicidal tendencies, history of abuse). More recently, Shaffer, Hartman, and Listwan (2009) found that moderate- to high-risk female drug court clients had significantly lower rates of recidivism than female nonparticipants, an effect that lasted over a 2-year period. This indicates that the drug court model is appropriate for women as well as men.

Sechrest and Shicor (2001) found race to be a significant predictor of drug court success in the Riverside County (California) Drug Court, with 68.9% of Caucasians graduating compared with 31.6% of African Americans and 42.1% of Hispanics. Other studies have reached similar conclusions, with Hartley and Phillips (2001) and Schiff and Terry (1997) proposing that this discrepancy could be due to cultural barriers and structural problems within drug courts. However, some studies have found no relationship between race and drug court graduation (Logan, Williams, Leukefeld, & Minton, 2000), or have found that race interacts with level of education to predict drug court success/failure (Butzin et al., 2002).

Education and Employment

Education and employment play an important role in drug court success. Educational attainment, typically measured by at least a high school education or GED, was one factor contributing to graduation in several studies (e.g., Brown, 2010; Brown, Allison, & Nieto, 2011; Deschenes, Ireland, & Kleinpeter, 2009). Being employed at the time of drug court enrollment is also associated with drug court success (e.g., Brown, 2010; Brown et al., 2011; Deschenes et al., 2009; Peters et al., 1999; Roll, Prendergast, Richardson, Burdon, & Ramirez, 2005). Peters et al. (1999) reported that 77% of graduates were employed full- or part-time upon enrollment compared with 54% of nongraduates.

Antisocial Personality Disorder (APD) and Previous Drug Treatment

APD and a history of drug treatment have been used to define high-risk drug court offenders (Festinger et al., 2002; Marlowe, Festinger, Lee, Dugosh, & Benasutti, 2006). This research suggests that drug court clients with these characteristics perform better in drug court if assigned to high-dose judicial status hearings, whereas low-risk clients without these characteristics performed equivalently regardless of the frequency of status hearings. Broadly, these findings suggest that higher risk offenders may require enhanced drug court supervision to complete drug court.

Present Study

The primary aim of this study was to determine whether baseline substance use severity—defined as lower scores on the ASI and/or baseline drug-negative urine screens—is associated with higher rates of drug court graduation. In addition, based on previous research on the characteristics of offenders likely to perform well or poorly in drug court, this study sought to create a model that predicted graduation. Measures included in the model were the ASI, baseline urine screen, socio-demographic variables, presence of APD, and previous drug treatment.

Method

Participants

Data were obtained from an archival database consisting of 284 adult drug offenders from three drug courts in Delaware. The drug courts are located in the urban city of Wilmington (New Castle County), the state capital of Dover (Kent County), and the rural community of Georgetown (Sussex County). Eligibility criteria for these programs required that defendants (a) be 18 years of age or older; (b) be charged with a misdemeanor drug offense involving possession or consumption of cannabis, possession of drug paraphernalia, possession of hypodermic syringes, or first-time driving under the influence; and (c) not have a history of an offense involving drug dealing or manufacturing, death or serious injury to a victim, or use of a deadly weapon. Depending on how crowded the drug court docket is, it may take 1 to 3 months after being arrested before offenders enter drug court.

All participants included in the database were originally obtained through random sampling from recent admissions to these preadjudication drug courts. Defendants are required to plead guilty to their charge(s) but the plea is held in abeyance until they graduate or are terminated from drug court. Upon graduation, charges are dropped and defendants’ arrest records are expunged. Termination results in a conviction of the charge(s). Of the 284 participants, 248 were participating in an experimental study examining the effects of status hearings on drug court outcomes, and the remaining 36 were participating in a “natural history” study examining standard drug court treatment. Only 251 clients were included in this sample as they had the status of being “graduated” or “terminated” from a drug court program (defined below); 33 clients were not included in the present sample because they were classified as either “long-term capias,” “charges dropped,” “deceased,” “unknown,” or “active.”

Procedure

ASI drug composite scores

Participants received a US$20 check for completing a baseline assessment battery administered by trained research technicians. In addition to being screened for APD, clients were administered the ASI-5 (McLellan et al., 1992). Several studies of ASI composite scores and lifetime items have yielded strong evidence of reliability, concurrent validity, predictive validity, and discriminative utility across groups of clients characterized by age, race, gender, and primary drug of abuse (Alterman et al., 1998; McLellan et al., 1992; McLellan et al., 1985; McLellan, Luborsky, O’Brien, & Woody, 1980). Baseline ASI drug composite scores range from 0.00 to 1.00. A drug composite score ≤0.04 is considered subthreshold for a drug use disorder, scores >0.04 but ≤0.24 are mild/moderate, and scores >0.24 are severe (Lee et al., 2001). These classifications were derived from comparison data on nonsubstance abuse clients from a Kaiser-Permanente Health Maintenance Organization (HMO; N = 9,398), outpatient substance abuse clients from the Drug Evaluation Network System (DENS; N = 2,707), and residential substance abuse clients from DENS (N = 5,256).

Baseline urine screen

Upon entry into the drug courts, each client provided a urine sample in the presence of a same-gender treatment staff member. Although this study focused on baseline urine screens, clients provided urine screens on a random weekly basis. Initial urine screens indicated whether participants had abstained from drugs during the time just before enrolling in drug court. The possible outcomes were drug-negative, drug-positive, or invalid. Drug screens were performed by an independent certified laboratory using enzyme multiple immunoassay technique (EMIT) and gas chromatography/mass spectrometry (GC/MS)confirmation of positive results for cannabis, alcohol, opiates, amphetamines, cocaine, phencyclidine (PCP), and other substances believed to be used by the individual. The urine specimens were tested for evidence of tampering or invalidity based on an analysis of creatinine, ph, and specific gravity in accordance with standard laboratory testing guidelines. In line with polices of the drug courts, invalid or tampered specimens were presumed to be drug-positive and the specimens were not credited as having been delivered as directed.

Drug court graduation

The outcome variable was measured dichotomously by whether participants graduated or were terminated from drug court. The three drug court programs are similar in structure and have almost identical graduation requirements. The programs are scheduled to be a minimum of 4 to 6 months in length, although most participants require 6 to 8 months to satisfy the conditions for graduation. To graduate, clients must complete a standard regimen of 12 weekly psychoeducational group-counseling sessions, attend weekly case management meetings, attend monthly judicial status hearings, provide at least 14 consecutive drug-negative urine specimens, remain arrest-free, obey program rules, and pay a US$200 court fee. The psychoeducational group sessions cover a standard sequence of prevention topics, including the pharmacology of drug and alcohol use, progression from substance use to dependence, the impact of addiction on the family, treatment options, HIV/AIDS risk reduction, and relapse prevention strategies. Participants provide urine specimens on a random, weekly basis through the drug court program.

“Termination” as used in this study refers to being expelled from the drug court program and criminally sentenced. Usually, clients are terminated from drug court programs for major infractions such as new arrests or having several consecutive missed or positive urine screens. Less-severe infractions (e.g., single positive urine screen) may result in clients taking longer to meet criteria for graduation but rarely result in formal termination.

Results

The mean age of the 251 participants was 24.53 (SD = 7.80), with a range of 18 to 56. The racial/ethnic breakdown included 144 Caucasian, 89 African American, and 18 Other. There were 195 males and 56 females. The average years of education completed was 11.80 (SD = 1.58). Regarding previous substance use treatment, 60 participants had received treatment, 187 had not, and 4 cases were unknown. There were 75 individuals who met the criterion for APD, 174 who did not, and 2 who were unknown. These participants were compared with the 33 participants who were excluded based on not having the status of “graduated” or “terminated” from drug court. Analyses revealed no statistically significant differences in age, race, gender, years of education, previous substance use treatment, and APD diagnosis.

Relationship Between Baseline ASI and Drug Court Graduation

A chi-square (χ²) test for independence was used to determine whether there was a consistent and predictable relationship between baseline ASI drug composite scores and graduation from drug court. The ASI drug composite score is measured on a continuum. Therefore, the cutoff scores of .04 and .24 were applied to separate subthreshold, mild to moderate, and severe levels of a drug use disorder (Lee et al., 2001). To further validate these cutoff scores, we conducted a Goodman and Kruskal’s gamma (G) correlation between the categorized ASI drug composite score and results from the baseline urine screen (drug-positive vs. drug-negative). A significant correlation was found, G(n = 182) = 0.37, p = .005.

The sample used for the chi-square analysis included 215 out of 251 participants because not all participants were administered the ASI by drug court staff upon entry into the drug court. Results of this analysis revealed that the proportion of participants graduating from drug court was 74.2% for those who scored in the subthreshold range (n = 97) on the ASI drug composite score, 60.0% for those who scored in the mild to moderate range (n = 105), and 38.5% for those who scored in the severe range (n = 13; see Figure 1). The difference in proportions was significant, χ²(2, n = 215) = 8.821, p = .012, Cramer’s V = 0.14 (small effect size).

Figure 1.

Proportions of those who graduated versus those who were terminated based on subthreshold, mild/moderate, and severe baseline ASI drug composite score.

To determine precisely where differences lie, follow-up chi-square tests compared each combination of ASI drug composite score classifications. A significant difference in graduation rates existed between the subthreshold and mild to moderate range, χ²(1, n = 202) = 4.604, p = .032, Cramer’s V = 0.15 (small), and the subthreshold and severe range, χ²(1, n = 110) = 6.983, p = 0.008, Cramer’s V = 0.25 (small). No significant difference in graduation rates existed when comparing the mild to moderate range with the severe range.

Relationship Between Baseline Urine Screen and Drug Court Graduation

A chi-square test for independence examined whether there was a relationship between baseline urine results (drug-positive vs. drug-negative) and drug court graduation. Participants whose baseline urine screen was invalid were excluded, so the resulting sample size was 216. The proportion of participants graduating from drug court was 85.9% for those who tested drug-negative (n = 92) and 64.5% for those who tested drug-positive (n = 124; see Figure 2), which reached statistical significance, χ²(1, n = 216) = 12.397, p < 0.001, Cramer’s V = 0.24 (small).

Figure 2.

Proportions of those who graduated versus those who were terminated based on urine drug screen at baseline.

Formulating a Model for Drug Court Graduation

Binary logistic regression was used to construct a model for drug court graduation. This analysis included 209 participants because of missing data on some variables. There were no statistically significant differences between these participants and those who were excluded. A nine-variable logistic model was created using the following variables: ASI drug composite scores, baseline urine screen results, previous treatment, presence of APD, age, race, gender, years educated, and drug court location. The last variable was added to control for differences in the drug courts from which participants were drawn. The variable “employment” was removed from the analysis because of an insufficient number of employed participants. “Graduation” was dummy-coded as “1” for those who graduated and “0” for those who were terminated. A summary of these variables is presented in Table 1.

Table 1.

Summary of Data for Logistic Regression.

Independent variables	Total (n = 209)	Graduated (n = 135)	Terminated (n = 74)
ASI
M	.08	.07	.10
SD	.08	.08	.09
Urine screen
Negative	79	66	13
Positive	99	62	37
Invalid	31	7	24
Age
M	24.38	24.65	23.89
SD	7.88	8.41	6.82
Race
Caucasian	115	87	28
African American	80	42	38
Other	14	6	8
Gender
Male	164	103	61
Female	45	32	13
Years educated
M	11.73	12.01	11.22
SD	1.52	1.52	1.39
APD
Yes	63	37	26
No	146	98	48
Previous treatment
Yes	51	28	23
No	158	107	51
Site
New Castle	154	94	60
Sussex	68	63	15
Kent	29	20	9

Note. ASI = Addiction Severity Index; APD = Antisocial Personality Disorder.

Evaluation of the logistic regression model

The logistic model proved to be a better fit to the data than the intercept-only model (i.e., model with no predictors), χ²(12) = 64.70, p < .001. The inferential goodness-of-fit test, the Hosmer–Lemeshow, yielded a nonsignificant χ²(8) value of 12.64, which suggests that the model is a good fit to the data. The accuracy of the model was 73.2%, sensitivity was 84.6%, and specificity was 52.1%. The observed and predicted frequencies for drug court graduation are shown in Table 2. Figure 3 displays an receiver operating characteristic (ROC) curve of the model’s predicted probabilitiesin which the area under the curve is .806 (p < .001), which suggests that the logistic regression was better able to classify participants as graduating versus terminating significantly better than chance. Finally, the two estimates of the variation explained (Cox and Snell R² = .264 and Nagelkerke R² = .363) suggest goodness of fit.

Table 2.

Observed and Predicted Frequencies for Graduation From Drug Court.

	Predicted		% Correct
Observed	No	Yes	% Correct
No	38	35	52.1
Yes	21	115	84.6
Overall % correct			73.2

Note. Sensitivity = 115 / (21 + 115)% = 84.6%. Specificity = 38 / (38 + 35)% = 52.1%. False positive = 35 / (35 + 115)% = 23.3%. False negative = 21 / (38 + 21)% = 35.6%.

Figure 3.

The area under the curve is .806 (p < .001) indicating the logistic regression classifies the group as graduating versus terminating better than chance.

Statistical tests of individual predictors

Results from the binary logistic regression regarding individual predictors are shown in Table 3. This analysis found that the baseline ASI drug composite score was a significant predictor of the probability of graduating from drug court. For every .01 decrease in the score, a drug court participant was .011 times more likely to graduate from drug court. In addition, those who had a baseline drug-negative urine specimen were more likely to graduate than those who tested drug-positive and those whose urine specimen was invalid. The odds of graduating were decreased by a factor of .354 for an individual with a baseline drug-positive urine screen rather than a baseline drug-negative urine screen. Similarly, the odds of graduating were reduced by a factor of .083 for a baseline invalid urine drug screen compared with a drug-negative urine screen.

Table 3.

Logistic Regression Analysis of 209 Drug Court Participants.

Predictor	b	SE	Wald’s χ²	df	p	Odds ratio
Constant	−1.285	1.619	0.629	1	.428	0.277
ASI drug score	−4.551	2.031	5.023	1	.025	0.011
Urine (0 = drug-negative, 1 = drug-positive)	−1.038	0.425	5.950	1	.015	0.354
Urine (0 = drug-negative, 1 = invalid)	−2.490	0.580	18.452	1	.000	0.083
APD diagnosis (0 = no, 1 = yes)	0.132	0.401	0.108	1	.743	1.141
Previous treatment (0 = no, 1 = yes)	−0.745	0.404	3.400	1	.065	0.475
Age	0.005	0.025	0.043	1	.836	1.005
Gender (0 = female, 1 = male)	−0.388	0.454	0.731	1	.393	0.679
Race (0 = Caucasian, 1 = Black)	−1.096	0.375	8.539	1	.003	0.334
Race (0 = Caucasian, 1 = Other)	−1.202	0.683	3.091	1	.079	0.301
Years educated	0.355	0.131	7.391	1	.007	1.426
Site (0 = New Castle, 1 = Sussex)	0.253	0.460	0.303	1	.582	1.288
Site (0 = New Castle, 1 = Kent)	−.382	0.701	0.297	1	.586	0.683

Note. ASI = Addiction Severity Index; APD = Antisocial Personality Disorder.

Race and years educated were also found to be significant predictors of graduation. The odds of graduation compared with termination were reduced by a factor of .334 for African Americans compared with Caucasian. However, this did not hold true when comparing Caucasians with the “Other” race category. Regarding education, for every 1 additional year of education, a drug court participant was 1.426 times more likely to graduate from drug court.

Discussion

Graduation rates in drug courts are cited as between 50% and 70% (Belenko, 1998, 2001; Belenko et al., 2007; General Accounting Office, 1997). Although drug courts assume that offenders are drug-addicted (NADCP, 1997), empirical evidence supports the proposition that many drug court offenders do not have a clinically significant substance use problem (DeMatteo et al., 2009). As such, it is not clear whether some drug court graduates actually respond to the intervention or simply have less severe drug problems upon entry into the drug court.

The primary goal of this study was to examine the relationship between baseline substance use severity and drug court graduation. The graduation rate in this study was similar to those found on a national level. The most important finding from this study was that measures of pre-drug court substance use severity, including baseline ASI drug composite scores and a single baseline urine screen, were significantly related to completion of drug court. Lower ASI drug composite scores were associated with higher rates of graduation. Specifically, those who scored in the subthreshold range had a significantly higher graduation rate than those who scored in the mild to moderate and severe ranges. The lack of difference between those who scored in the mild to moderate and severe ranges may be due to low statistical power. Drug-negative baseline urine screens were also associated with higher graduation rates. Thus, while overall graduation rates were high, drug use severity prior to drug court entry may be a predictor of success or failure in the drug court. The ability to stop drug use prior to entering a drug court may be indicative of success in abstaining from drugs during the course of drug court, which may suggest a less entrenched drug problem that lacks compulsivity and other signs of drug dependence. Subsequently, these clients are more likely to pass random weekly urine screens, complete treatment, acquire positive rewards for program accomplishments, and ultimately graduate. Therefore, one could argue that drug courts could increase their efficiency by reducing the intensity of services for these clients and possibly maintain or increase the intensity for those who enter a drug court with higher drug use severity.

Other than the baseline substance use, factors associated with graduation include education, employment, demographic characteristics, previous treatment, personality style, and motivation. Therefore, the present study integrated other variables previously researched in the context of drug courts, including socio-demographic variables, presence of APD, and prior treatment. Although race and years educated were predictive of drug court graduation, age and gender were not. The findings regarding gender, race, and years educated are consistent with previous research, while the lack of a relationship between age and graduation was somewhat at odds with prior research (Butzin et al., 2002; Saum et al., 2001; Schiff & Terry, 1997; Sechrest & Shicor, 2001). In this study, 75% of the participants were between the ages of 18 and 26. The restricted age range may account for this variable not being a significant predictor of drug court graduation. Finally, APD and previous treatment were incorporated into the analysis because these variables may be attributed to high-risk offenders one would expect to be less successful in drug court. However, these features by themselves were not related to drug court graduation in this analysis.

This study provides additional information on what could be going on inside the “drug court black box” (Goldkamp et al., 2001) such that some participants are able to successfully complete the program because of a low substance use severity at baseline. This has several implications. One implication is that drug courts would benefit from tailoring interventions to low-needs versus high-needs offenders based on substance use severity. A national survey of drug courts found that a large majority of drug courts use a standard protocol of clinical services: 12-step groups/relapse prevention (93%), education remediation (92%), mental health treatment (91%), anger management (87%), vocational training (86%), and parenting education (84%) (Peyton & Gossweiler, 2001). The participants in the current study received similar services: psychosocial group counseling (12 weekly sessions), weekly case management meetings, and weekly random urine drug screens, in addition to attending judicial status hearings on a monthly basis. This potentially “one-size-fits-all” approach for drug courts may be inefficient and not cost-effective. Research has shown that graduation rates of high-risk offenders improve when they are matched to differential dosages of judicial supervision (as opposed to the usual schedule of hearings) based on clinical assessments of their needs (Marlowe, Festinger, Dugosh, Lee, & Benasutti, 2007). In the same manner, drug courts should minimally evaluate current treatments including pharmacotherapies and psychosocial interventions provided to high-needs clients. Much like high-risk clients with APD and previous drug treatment, higher baseline substance use severity, as measured by the ASI, and an initial urine screen may warrant more frequent status hearings and increased supervision and treatment

Low-needs clients should require less resources in terms of supervisory and treatment services. One might wonder whether low-needs clients simply do not require drug court intervention, but disparate recidivism rates between drug court graduates and individuals in the traditional courts suggest otherwise (Belenko et al., 2007; Government Accountability Office, 2011; Wilson et al., 2006). Thus, individuals who have not yet developed a clinical syndrome but are in danger of acquiring future addictive behaviors are ideal targets for secondary prevention strategies (DeMatteo, 2010). Such services could assist low-needs drug offenders to avoid risky situations that may lead to the acquisition of addictive behavior. For instance, one intervention would be to have them develop daily or weekly activities that decrease involvement with the “people, places, and things” that bring about drug use and increase involvement with drug-incompatible peers and activities (see DeMatteo, 2010). On the other hand, given the impressive graduation rates for offenders with low substance use severity found in this study, one could also argue that although the interventions were appropriate for these clients, they could be more intensive for high-needs offenders. Although potentially true, it is possible that similar success rates could be achieved more efficiently through the implementation of less-intensive secondary interventions.

Another implication is the need for more in-depth screening to exclude those without a minimal identifiable disorder. Drug courts may provide certain advantages such as avoiding incarceration, avoiding or reducing probationary obligations, receiving a sentence of “time served,” or having one’s arrest record expunged. Thus, avoidance of more serious criminal consequences may be an incentive for potential drug court participants to exaggerate or even feign a drug problem to be admitted into drug court.

This study has several limitations. The generalizability of findings is limited because data were collected from only three drug courts in Delaware. However, the locations of these drug courts were in urban and rural areas, which could provide a relatively diverse sample. Findings may also have limited generalizability because all three programs were misdemeanor drug courts. Possession or consumption of cannabis accounted for a substantial portion of the offenses represented by the sample, which may have limited the number of high-needs clients included in the study. As the average length of time between arrest and entrance into drug court was 1 to 3 months, this could have potentially given drug users the necessary time to eliminate substances from their system so that the baseline urine screen would be negative. In effect, this could have also reduced the number of high-needs clients in the study. Moreover, while the ASI comes with strong evidence of reliability and validity, the use of cutoff scores, specifically for the ASI drug composite score, has not been well established in previous research. These cutoff scores, acquired from Lee et al. (2001), separated those with subthreshold, mild to moderate, and severe substance use problems. No subsequent discussion was found regarding the sensitivity and specificity of ASI drug composite cut scores. However, the present study may be viewed as further establishing these cutoffs.

There are several directions for future research. One is the replication of this study with a different or more diverse sample; specifically, a sample likely to have more individuals with high baseline drug use severity. Replications may also want to incorporate additional variables that the current study did not include because of the lack of variability, such as employment on drug court entry, marital status, and the drug of choice. Additional measures of drug use severity other than the ASI and urine screen should be implemented as well to increase the validity of findings. Similarly, the predictive model produced by the current analysis should be cross-validated in other drug courts to estimate how accurately the model predicts graduation. Future researchers should consider using a split-sample to develop and subsequently cross-validate the predictive model.

More broadly, future research should examine treatments currently used in drug court programs with the goal of understanding which treatments provide the most benefit for specific types of offenders. For example, treatment dosage should vary depending on baseline drug use severity as well as risk/needs levels in general, with high-risk and high-needs individuals receiving greater dosages. Possible less-intensive and less-costly secondary interventions for low-risk and low-needs drug court clients should also be explored. Tailoring and adapting interventions to the specific client’s needs could foster greater efficiency as well as increased graduation rates from drug courts. This in turn could yield two important public policy outcomes: more effective use of allocated funding and reduced criminal recidivism.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Sanjay Shah, JD, PhD, received his doctoral training from Drexel University Department of Psychology and is a current postdoctoral fellow at the Emory University School of Medicine. His research interests include forensic mental health assessment, diversion from standard prosecution, and forensic neuropsychology.

David DeMatteo, JD, PhD, is an associate professor of psychology and law, and director of the JD/PhD program in law and psychology at Drexel University. His research interests include drug-involved offenders, psychopathy, forensic mental health assessment, and diversion from standard prosecution.

Michael Keesler, JD, MS, is a doctoral graduate student at Drexel University Department of Psychology. His current research interests include capital litigation, psychopathy, forensic neuropsychology, and jury decision making.

Jennie Davis, MS, completed her master’s training at Drexel University Department of Psychology and is currently a student at Boston College Law School. Her research interests include jail diversion, specialty courts, capital sentencing, the probative value of mental health evidence and eyewitness and expert testimony, and veterans issues.

Kirk Heilbrun, PhD, is currently professor, Department of Psychology, Drexel University, and codirector of the Pennsylvania Mental Health and Justice Center of Excellence. His current research focuses on juvenile and adult offenders, legal decision making, forensic evaluation associated with such decision making, and diversion.

David S. Festinger, PhD, is the director of TRI’s Section on Law and Ethics Research and an adjunct assistant professor of psychiatry at the University of Pennsylvania School of Medicine. His research has focused primarily on bringing experimental research methods to bear on major ethical questions facing research participants in substance abuse research.

References

Alterman

A. I.

Kampman

Boardman

C. R.

Cacciola

J. S.

Rutherford

M. J.

McKay

J. R.

Maany

(1997). A cocaine-positive baseline urine predicts outpatient treatment attrition and failure to attain initial abstinence. Drug and Alcohol Dependence, 46, 79-85.

Alterman

A. I.

McDermott

P. A.

Cook

T. G.

Metzger

Rutherford

M. J.

Caccola

J. S.

Brown

L. S.

Jr. (1998). New scales to assess change in the Addiction Severity Index for the opiod, cocaine, and alcohol dependent. Psychology of Addictive Behaviors, 2, 233-246.

Alterman

A. I.

McKay

J. R.

Mulvaney

F. D.

McLellan

A. T.

(1996). Prediction of attrition from day hospital treatment in lower socioeconomic cocaine-dependent men. Drug and Alcohol Dependence, 40, 227-233.

Belenko

(1998). Research on drug courts: A critical review. National Drug Court Institute Review, 1, 1-42.

Belenko

(1999). Research on drug courts: A critical review: 1999 update. National Drug Court Institute Review, 2, 1-58.

Belenko

(2001). Research on drug courts: A critical review: 2001 update. New York, NY: National Center on Addiction and Substance Abuse at Columbia University.

Belenko

(2002). Drug courts. In Leukefeld

C. G.

Tims

Farabee

(Eds.), Treatment of drug offenders: Policies and issues (pp. 301-318). New York, NY: Springer.

Belenko

DeMatteo

Patapis

(2007). Drug courts. In Springer

D. W.

Roberts

A. R.

(Eds.), Handbook of forensic mental health with victims and offenders: Assessment, treatment, and research (pp. 385-423). New York, NY: Springer.

Belenko

Patapis

French

M. T.

(2005). Economic benefits of drug treatment: A critical review of the evidence for policy makers. Philadelphia, PA: Treatment Research Institute.

10.

Bhati

A. S.

Roman

J. K.

Chalfin

(2008). To treat or not to treat: Evidence on the prospects of expanding treatment to drug-involved offenders. Washington, DC: Urban Institute.

11.

Brown

R. T.

(2010). Associations with substance abuse treatment completion among drug court participants. Substance Use & Misuse, 45, 1874-1891.

12.

Brown

R. T.

Allison

P. A.

Nieto

F. J.

(2011). Impact of jail sanctions during drug court participation upon substance abuse treatment completion. Addiction, 106, 135-142.

13.

Butzin

Saum

Scarpitti

(2002). Factors associated with completion of a drug treatment court diversion program. Substance Use & Misuse, 37, 1615-1633.

14.

DeMatteo

(2010). A proposed prevention intervention for nondrug-dependent drug court clients. Journal of Cognitive Psychotherapy, 24, 104-115.

15.

DeMatteo

Marlowe

D. B.

Festinger

D. S.

Arabia

P. L.

(2009). Outcome trajectories in drug court: Do all participants have drug problems? Criminal Justice and Behavior, 36, 354-368.

16.

Deschenes

E. P.

Ireland

Kleinpeter

C. B.

(2009). Enhancing drug court success. Journal of Offender Rehabilitation, 48, 19-36.

17.

Ehrman

R. N.

Robbins

S. J.

Cornish

J. W.

(2001). Results of a baseline urine test predict levels of cocaine use during treatment. Drug and Alcohol Dependence, 62, 1-7.

18.

Festinger

D. S.

Marlowe

D. B.

Lee

P. A.

Kirby

K. C.

Bovasso

McLellan

A. T.

(2002). Status hearings in drug court: When more is less and less is more. Drug and Alcohol Dependence, 68, 151-157.

19.

General Accounting Office. (1997). Drug courts: Overview of growth, characteristics, and results. Washington, DC: Author.

20.

Goldkamp

J. S.

White

M. D.

Robinson

J. B.

(2001). Do drug courts work? Getting inside the drug court black box. Journal of Drug Issues, 31, 27-72.

21.

Government Accountability Office. (2011). Adult drug courts. Washington, DC: Author.

22.

Hartley

R. E.

Phillips

R. C.

(2001). Who graduates from drug courts? American Journal of Criminal Justice, 26, 107-119.

23.

Hickert

A. O.

Boyle

S. W.

Tollefson

D. R.

(2009). Factors that predict drug court completion and drop out: Findings from an evaluation of Salt Lake county’s adult felony drug court. Journal of Social Science Research, 35, 149-162.

24.

Higgins

S. T.

Budney

A. J.

Bickel

W. K.

Hughes

J. R.

Foerg

Badger

(1993). Achieving cocaine abstinence with a behavioral approach. American Journal of Psychiatry, 150, 763-769.

25.

Jones

C. G. A.

Kemp

R. I.

(2011). The relationship between early-phase substance-use trajectories and drug court outcomes. Criminal Justice and Behavior, 38, 913-933.

26.

Koob

Brocato

Kleinpeter

(2011). Enhancing residential treatment for drug court participants. Journal of Offender Rehabilitation, 50, 252-271.

27.

Lee

P. A.

Marlowe

D. B.

Festinger

D. S.

Cacciola

J. S.

McNellis

Schepise

N. M.

McLellan

A. T.

(2001). Did “Breaking the Cycle” (BTC) clients receive appropriate services? [abstract]. Drug and Alcohol Dependence, 63(Suppl. 1), S89.

28.

Logan

Williams

Leukefeld

Minton

(2000). A drug court process evaluation: Methodology and findings. International Journal of Offender Therapy and Comparative Criminology, 44, 369-394.

29.

Marlowe

D. B.

Festinger

D. S.

Arabia

P. L.

Dugosh

K. L.

Benasutti

K. M.

Croft

J. R.

McKay

J. R.

(2008). Adaptive interventions in drug court: A pilot experiment. Criminal Justice Review, 33, 343-360.

30.

Marlowe

D. B.

Festinger

Dugosh

Lee

Benasutti

(2007). Adapting judicial supervision to the risk level of drug offenders: Discharge and 6-mounth outcomes from a prospective matching study. Drug and Alcohol Dependence, 88, S4-S13.

31.

Marlowe

D. B.

Festinger

Lee

(2003). The role of judicial status hearings in drug court. Offender Substance Abuse Report, 3, 33-46.

32.

Marlowe

D. B.

Festinger

Lee

(2004). The judge is a key component of drug court. Drug Court Review, 4(2), 1-34.

33.

Marlowe

D. B.

Festinger

Lee

Dugosh

Benasutti

(2006). Matching judicial supervision to clients’ risk status in drug court. Crime & Delinquency, 52, 52-76.

34.

Marlowe

D. B.

Patapis

DeMatteo

(2003). Amenability to treatment of drug offenders. Federal Probation, 67(2), 40-46.

35.

McLellan

A. T.

Cacciola

Kushner

Peters

Smith

Pettinati

(1992). The fifth edition of the Addiction Severity Index: Cautions, additions and normative data. Journal of Substance Abuse Treatment, 9, 461-480.

36.

McLellan

A. T.

Luborsky

Cacciola

J. S.

Griffith

Evans

Barr

H. L.

O’Brien

C. P.

(1985). New data from the Addiction Severity Index: Reliability and validity in three centers. The Journal of Nervous and Mental Disease, 173, 412-423.

37.

McLellan

A. T.

Luborsky

O’Brien

C. P.

Woody

G. E.

(1980). An improved diagnostic instrument for substance abuse clients: The Addiction Severity Index. The Journal of Nervous and Mental Disease, 168, 26-33.

38.

Mitchell

Wilson

D. B.

Eggers

MacKenzie

D. L.

(2012). Assessing the effectiveness of drug courts on recidivism: A meta-analytic review of traditional and non-traditional drug courts. Journal of Criminal Justice, 40, 60-71.

39.

National Association of Drug Court Professionals. (1997). Defining drug courts: The key components. Washington, DC: Office of Justice Programs, U.S. Department of Justice.

40.

Peters

Haas

Murrin

(1999). Predictors of retention and arrest in drug courts. National Drug Court Institute Review, 2, 33-60.

41.

Peyton

E. A.

Gossweiler

(2001). Treatment services in adult drug courts: Report on the 1999 National Drug Court Treatment Survey, executive summary. Washington, DC: U.S. Department of Justice.

42.

Roll

J. M.

Prendergast

Richardson

Burdon

Ramirez

(2005). Identifying predictors of treatment outcome in a drug court program. American Journal of Drug and Alcohol Abuse, 31, 641-656.

43.

Rossman

Roman

Zweig

Rempel

Lindquist

(2011). The multi-site adult drug court evaluation: What’s happening in drug courts? A portrait of adult drug courts in 2004 (Submitted to the National Institute of Justice, U.S. Department of Justice). Washington, DC: The Urban Institute.

44.

Saum

Scarpitti

Robbins

(2001). Violent offenders in drug court. Journal of Drug Issues, 31, 107-128.

45.

Schiff

Terry

(1997). Predicting graduation from Broward County’s dedicated drug treatment court. Justice System Journal, 19, 291-310.

46.

Sechrest

D. K.

Shicor

(2001). Determinants of graduation from a day treatment drug court in California: A preliminary study. Journal of Drug Issues, 31, 129-148.

47.

Shaffer

D. K.

(2011). Looking inside the black box of drug courts: A meta-analytic review. Justice Quarterly, 28, 493-521.

48.

Shaffer

D. K.

Hartman

J. L.

Listwan

S. J.

(2009). Drug abusing women in the community: The impact of drug court involvement on recidivism. Journal of Drug Issues, 39, 803-827.

49.

Walsh

(2011). Addicted to courts: How a growing dependence on drug courts impacts people and communities. Washington, DC: Justice Policy Institute.

50.

Wilson

Mitchell

Mackenzie

(2006). A systematic review of drug court effects on recidivism. Journal of Experimental Criminology, 2, 459-487.