THE AMERICAN JOURNAL OF DRUG AND ALCOHOL ABUSE
Vol. 30, No. 3, pp. 605–625, 2004
Treatment Retention Predictors of Drug Court
Participants in a Rural State
Allison Mateyoke-Scrivner, M.S.,*J. Matthew Webster,
Michele Staton, and Carl Leukefeld
Center on Drug and Alcohol Research, University of Kentucky,
Lexington, Kentuky, USA
Factors distinguishing clients who complete drug court treatment from
those who do not complete drug court have been documented, but
differences between urban and rural drug court participants have not
been examined. The present study focuses on examining mental health,
drug use, criminal activity, and education/employment as factors that are
associated with treatment retention, which is measured by graduation
from a rural and urban drug court. Study findings indicate that for the
urban drug court, marital status, employment, drug use, and criminal
activity predicted graduation. For the rural drug court, however,
graduation was only predicted by age and juvenile incarceration.
Findings from this study suggest there are different factors associated
*Correspondence: Allison Mateyoke-Scrivner, M.S., Center on Drug and Alcohol
Research, University of Kentucky, 643 Maxwelton Ct., Lexington, KY 40506-0350,
USA; E-mail: email@example.com.
Copyright D 2004 by Marcel Dekker, Inc.
0095-2990 (Print); 1097-9891 (Online)
with drug court retention/graduation between urban and rural drug
court settings. It is suggested that drug court administrators and other
could use this information to better assess potential participants and to
Drug court; Urban; Rural; Treatment retention;
According to the Bureau of Justice Statistics, the number of jail
inmates who were regular drug users or were incarcerated for a drug
offense increased substantially in 1998 from 261,000 eleven years earlier to
over 400,000 (1). America’s ‘‘war on drugs’’ during the 1980s also
increased the number of drug arrests as a result of mandatory sentencing
practices for drug offenders. This dramatic increase in arrests created
overcrowding in the state and federal justice system as well as correctional
systems. One of the more successful attempts to reduce this overcrowding
in the last two decades are drug courts.
Drug courts view the increase in drug use/arrests as a criminal justice
issue as well as a public health problem that can be remedied (2). Drug
courts were recognized as a way to alleviate overcrowding and provided
other benefits which include: 1) reduced recidivism; 2) decreased drug use;
3) increased birth rates of drug-free babies; 4) greater access to mentor
groups and community resources; 5) increased efforts by participants at
long-term relapse prevention efforts; and 6) cost-effective treatment (3–5).
By June 2001, a total of 226,000 individuals were enrolled in adult drug
court programs in the United States with an estimated 74,000 graduates (6).
Treatment Retention and Outcomes
Successful completion of substance abuse treatment, including drug
court treatment, leads to a higher likelihood of a continued drug-free
lifestyle (7), as well as decreased criminal activity, greater psychological
functioning (i.e., lower rates of depression), and increased employment
(8,9). Clients who drop out early or are terminated from treatment are more
likely to relapse and have future increased legal and employment problems
(7). Furthermore, drug court clients who stay in treatment for one year or
longer are five times more likely to have better outcomes than those who
drop out or participate for a minimum time period (10). Treatment also can
improve relationships with friends, family and employers (11) which, in
turn, lead to decreased substance use and criminal involvement (12–14). In
606 Mateyoke-Scrivner et al.
addition, drug court graduates are less likely to recidivate and use fewer
substances after treatment than nontreated offenders, and if graduates do
reoffend, they do so after a longer period of time than do those who do not
Factors related to treatment retention have been identified. Although
sometimes conflicting, distinctions have been noted between treatment
completers and dropouts. For example, one study found that treatment
completers were more likely than were dropouts to have better social
connections, more close friends, and greater social conformity (7). In
addition, completers were less likely to report having significant problems
with a spouse or significant other in the 30 days prior to arrest, while
having such problems with a significant other increased the odds of dropout
(16). Similarly, being married predicted treatment retention (9,17–20).
However, differences between completers/noncompleters in urban and rural
areas, as well as those who were most likely to benefit from the drug court
program, have not been fully explored (3).
While some studies indicate that females are more likely than are
males to remain in treatment (21,22), other studies have reported no
statistical difference between genders (23,24) or reported that women were
more likely to drop out of treatment than were men (9). Race/ethnicity has
also been examined as a factor in treatment retention. For example, a recent
study found that only one-third (31.6%) of blacks completed treatment
compared to two-third (68.9%) of whites (25), while another study reported
black women had a lower chance of treatment completion than did other
women (17). Conversely, other studies indicate there are no significant
differences in the rates of treatment nonsuccess between races (9,23,24).
Two demographic characteristics, which are related to one another as
well as to retention, are education and employment. Specifically, a low
educational level and/or being unemployed have consistently predicted
relapse and treatment dropout (7,16,18,20,25–27). According to Sayre, etal.
(16), clients with lower education levels have more difficulty expressing
their needs, completing treatment assessments, may feel inferior to those
participants with more education, and are more apt to drop out. Moreover,
those individuals who cannot find stable employment because of limited
education are less likely to resist temptation to use drugs after treatment (8).
Studies also suggest that employment stability is associated with reduced
substance use, severity of relapse, improved community functioning, and
community reintegration (28–32). In particular, stable employment interrupts
addiction patterns and unemployment has proven to be a stronger predictor of
relapse than the severity of a client’s addiction (30).
In addition to employment and education, another demographic
variable that has been associated with treatment completion is age.
Specifically, older clients are more likely to complete treatment with
Retention Predictors 607
positive outcomes than are younger clients (17,24,25,27). It is possible that
older participants with a longer history of drug use are more ready for
change than are younger clients who have not used as long (23). In
addition, older clients may have more stable lifestyles, which prompt
change: and/or they may have just ‘‘grown out’’ of their drug use.
According to the maturation hypothesis, addicts become inactive because of
their lifecycle (that is, addicts stop using drugs in their 30s) or due to the
number of years of their addiction. For example, Winick (1962) explained
that individuals in their teens and late 20s begin using drugs to cope with
social pressures, decisions, and as an expression of social needs. By their
30s, addicts feel less is expected of them. There are not as many pressures,
stressors, and strains as before, thus establishing more stable emotions (33).
In addition to demographic characteristics, treatment completers and
noncompleters can also differ in terms of their psychological behavior, drug
use, and criminal activity. For example, several studies reported that mental
health problems are associated with substance use. One study in particular
reported that over one-third of those receiving mental health treatment had
substance abuse problems that influenced their current mental health status
(35,36). Psychiatric disorders such as anxiety and depression are among the
most frequently reported problems by substance abusers (36). Although there
are contradictory findings, one study reported that treatment noncompleters
were four times more likely to have psychiatric disorders than were those
who completed treatment (7). Psychiatric status has been related to treatment
completion/noncompletion, as well as posttreatment outcomes (9).
Clients with longer and more severe drug use histories at treatment
entry are less likely to complete treatment (7). There are also data that
suggest that clients with both cocaine dependence and alcohol dependence
may have more severe dependency issues and lower rates of treatment
completion (37). Furthermore, findings from a study by Brown, Voskul, and
Leyman (1977) suggest that the primary drug of choice was different for
urban and rural areas. While opiates were reported to be the primary drug
among urban users, rural users preferred marijuana or inhalants (38). An
analysis of the data from the National Household Survey found that rural
respondents reported using cocaine and marijuana less frequently than did
urban respondents, which could be related to the limited availability of
cocaine in rural areas (34). Another study reported that alcohol abuse was
more prevalent in rural areas when compared to urban areas (39). Residents
of rural areas are generally older and, therefore, treatment needs in rural
areas are different than those in urban areas (34). Research has shown that,
despite the substance use, there is a higher level of disapproval for drugs
and alcohol use in rural areas than urban areas (40), which may account for
the mental health problems rural drug users face and possibly the different
drug treatment needs (34).
608 Mateyoke-Scrivner et al.
Studies on the relationship between treatment retention and criminal
activity reveal that early involvement with the criminal justice system and
having a more extensive criminal history produce less favorable treatment
outcomes (27,41). In fact, it has been reported that each felony conviction
increases the likelihood of dropping out of treatment (7). While the number
of drug-specific charges do not seem to predict treatment dropout, clients
with more charges before treatment and those with a history of violence
were more likely to have negative treatment outcomes and pretreatment
dropout (24,43). Other studies, however, have found that clients with a
higher number of prior offenses were more likely to complete treatment or
remain in treatment longer (16,19,25). Furthermore, a greater number of
clients who completed treatment were involved in trafficking illegal
The Bureau of Justice Statistics reported that almost three-fourths
(74%) of offenders with 11 or more prior arrests will be rearrested within
three years; however, treatment program graduates with 11 or more
offenses are half as likely to reoffend when compared to noncompleters
(44). In rural areas, however, activities that may be reported as being
criminal in urban areas may not be considered a crime in rural areas
because of the rural ‘‘culture’’ of keeping issues within the community,
handling problems informally, or the communities’ limited resources (i.e.,
medical examiners in cases to death) (45). Consequently, rural clients may
appear to have committed fewer crimes, but may actually have committed
the same number of crimes as urban clients.
Kentucky Drug Courts
According to the Census Bureau, a rural area is one that has a
population of less than 50,000 residents combined with its adjacent areas.
Kentucky is one of 15 states in which more than half of its residents live in
areas with less than 25,000 people (46). There are 108 counties in Kentucky
which, based on this definition, are rural and 12 are urban. Overall, rural
residents constitute one-fifth of the U.S. population and one-third of the
nation’s persons in poverty (47). Kentucky’s rural areas have lower
education levels (40) with the number of persons with a high school degree,
well below the national average. Furthermore, Kentucky’s poverty rate is
higher than the national average (16% vs. 13%) (48,49).
Kentucky introduced its first drug court in 1993 for addicted,
nonviolent offenders. By January 2001, there were 10 Kentucky adult
drug courts and 20 others in the planning stages (50,51). One of the largest
drug courts in Kentucky is operated in Fayette County. The Fayette County
Drug Court was established by the Administrative Office of the Court in
1996 and serves an urban population (51). The program includes three
phases which take 12 to 24 months to complete depending on individual
progress. Participants are required to pay appropriate restitution, find court
approved housing, and maintain full-time employment or be involved in
school, most frequently GED coursework.
At the end of 2001, the Fayette County Drug Court had 491
participants, 146 graduates, and a 48% retention rate (6). The majority of
the Fayette County Drug Court clients were male (71%) and African
American (64%) with an average age of 31 (51). Those who graduated from
drug court were more likely to be older; however, the difference in ages
between completers and noncompleters is diminishing (51).
Drug courts in Kentucky, however, are not limited to larger, more
urban areas. Several drug courts have been implemented and are being
planned in rural areas. For example, the Warren County Drug Court was
established in 1997 and serves a community of less than 50,000 people
when compared to Fayette County, which serves a population of approxi-
mately 260,000 according to the 2000 census. Graduates of the Warren
County Drug Court were more likely to be male, white, and older (51). Unlike
Fayette County, which had a retention rate of 48%, the Warren County Drug
Court had a retention rate of 80% (6). This difference in retention between an
urban and rural drug court warrants further research.
Research on crime, drugs, and treatment has primarily focused on
urban areas (45). Because there is limited research that examines both rural
and urban drug courts, the current study focuses on drug court treatment
retention in rural and urban areas using data collected from the Enhancing
Drug Court Retention in a Rural State project, which is supported by the
National Institute on Drug Abuse. Specifically, mental health, drug use,
criminal activity, and education/employment will be examined to better
understand differences between rural and urban drug courts clients. This
study will provide information to fill gaps in the literature on drug court
treatment in rural as well as urban areas and to provide possible
explanations for the differences in rural and urban drug court retention.
The 500 participants for the current study were selected from two
Kentucky drug court sites in order to provide urban and rural comparisons.
610 Mateyoke-Scrivner et al.
The Fayette County Drug Court program was selected as an urban setting
because it is located at the crossroads of two major interstates, and the area
has a reputation as a marijuana producer. The Warren County Drug Court
was selected as a rural site because it is a rural county with a rising drug
problem. Study eligibility was based on criteria for entry into the drug
court program including nonviolent charges, a self-admitted drug problem,
urine test consent, and an Addiction Severity Index classification as a
Most of the study participants (69.0%) were male, with an average age
of 30.4 years (range 18–57). Over half of the participants were white
(59.0%), with an average of 11.8 years of education. In addition,
participants were more likely to have never been married (56.0%), with
1.6 children (range 0–21). Participants included 50% (N = 250) from the
Fayette County Drug court (urban site) and 50% (N = 250) from the Warren
County Drug Court (rural site).
Participants were recruited for the Enhancing Drug Court Retention
project between March 2000 and December 2002 when they appeared
before a drug court judge. Project participation was voluntary and subjects
interested in participating were interviewed face-to-face within two weeks
of drug court program entry after receiving their informed consent. At the
time of interview, participants were given a description of the project,
procedures, and assurance of confidentiality following International Review
Board (IRB)-approved informed consent procedures. Each question was
individually asked and recorded by the interviewer. The interview lasted
approximately two hours and all participants completing an interview were
paid for their participation.
The interview instrument contained measures of lifetime, past year, and
30-day time periods focusing on the time before drug court entry. The
specific measures included:
The Addiction Severity Index (ASI), which is a reliable measure of
drug and alcohol abuse severity, health, and treatment change through
demographic information and personal histories of health, mental status,
legal status, family and social relationships, and employment (52,53). For
the present study, the ASI was used to examine self-reported histories of
drug use, mental health, and treatment. Drug use histories were obtained by
asking participants how many years and how often they regularly used
alcohol, marijuana, cocaine, inhalants, amphetamines, methamphetamines,
opiates, heroin, and multiple drugs. Pattern of drug usage was coded from 0
(never) to 7 (about 4 or more times daily).
Criminal history was measured with each consenting participant’s
criminal record from their drug court file. Overall, over 95% of the total
sample—97% from Fayette County and 94% from Warren County—agreed
to have their criminal records included in the study. The information was
categorized and summed to obtain the total number of charges, total
number of convictions, and first and last dates of arrest.
The Brief Symptom Inventory was developed from the SCL-90-R
instrument and was designed to evaluate psychological symptoms at intake.
This inventory includes nine subscales: Somatization, Obsessive-Compul-
Paranoid Ideation, and Psychoticism. A Global Severity Index was created by
summing the responses on each subscale (54).
Variables selected for analysis, for those who had graduated or
terminated drug court (N = 347), included: 1) demographics; 2) employment
and income; 3) drug use; 4) criminal history. Data were analyzed in a series
of analysis of variance (ANOVA) and chi-square tests with drug court status
(graduated or terminated) and rural/urban as the fixed factors. Demographic
variables were analyzed by comparing graduated (N = 132) vs. terminated
(N = 215), and again by urban (N = 181) vs. rural (N = 166). Subjects were
then categorized into one of four groups: graduated urban, graduated rural,
terminated urban, and terminated rural. After the initial ANOVA test was
conducted, the Tukey Honestly Significantly Different (HSD) post-hoc
comparison procedure was applied to test for significant differences between
the four status groups. After significance was determined, significant var-
iables were used as predictors in a logistic regression equation to determine
which factors predicted graduation or termination for the entire sample and
then again on both the urban and rural samples separately.
Graduated vs. Terminated
The ANOVA and chi-square tests revealed significant differences be-
tween terminated and graduated groups and between urban and rural groups.
612 Mateyoke-Scrivner et al.
Significant effects between the graduated and terminated groups were found
for age, F(1, 345) = 10.65, p < .001; marital status w2(1, N = 346) = 3.32,
p < .10; employment since drug court entry w2(2, N = 326) = 10.2; p < .01;
and years of education F(1, 345) = 9.70, p < .01. These differences indicated
that participants who graduated from drug court were slightly older
(M = 31.80, SD = 9.41), were more educated (M = 12.17, SD = 1.85), were
more likely to be employed full-time, and were less likely to be married.
Urban vs. Rural
For urban and rural, significant main effects were found for race, w2
(1, N = 347) = 13.69, p < .001, and employment since drug court entry, w2
(2, N = 326) = 16.9, p < .001. Specifically, Warren County Drug Court
participants were more likely to be white, while urban participants were
more likely to be employed full-time. The number of children was
significantly different between the urban and rural participants F(1,
340) = 3.49, p < .10, with urban participants having more children
(M = 1.7, SD = 1.74) than rural participants (M = 1.4, SD = 1.46). In
addition, rural participants reported slightly more legal income in the six
months prior to drug court entry F(1, 345) = 3.68, p < .06) (M = $4,518.84,
SD = 5344.41) than did urban clients (M = $3,475.68, SD = 4781.85).
Graduation Status by Urban/Rural
While no significant differences were found for income or mental
health problems among the four groups, significant differences were found
for drug use (e.g., lifetime drug use for each substance, pattern of usage;
see Table 1). Terminated urban participants used cocaine and multiple
substances more often (M = 4.3, SD = 2.73 for cocaine; M = 4.2, SD = 2.56
for multiple substances) and for more years (M = 5.7, SD = 5.73 for cocaine;
M = 6.4, SD = 6.13 for multiple substances) than terminated rural partic-
ipants (M = 1.7, SD = 2.45, for pattern of cocaine usage; M = 2.8, SD = 2.71
for pattern of multiple substances; M = 1.9, SD = 3.96 for number of years
of cocaine use; M = 3.3, SD = 5.11 for number of years of multiple
substance use). In addition, urban graduates used cocaine (M = 4.5,
SD = 6.56) and multiple substances (M = 5.7, SD = 6.06) for a longer period
of time than did rural participants who were terminated from the drug court
program (M = 1.9, SD = 3.96 for cocaine use; M = 3.3, SD = 5.11 for
multiple substance use). Also, terminated urban participants used cocaine
more often than did urban graduates (M = 4.3, SD = 2.73 and M = 3.0,
SD = 2.75, respectively).
Retention Predictors 613
Demographic information as a function of graduation status and
UrbanRuralUrbanRural F (df)
Pattern of usage
4.63 (3, 343)
3.62 (3, 343)
1.33 (3, 343)
1.35 (3, 342)
10.51 (3, 342)
6.56 (3, 343)
3.41 (3, 343)
1.57 (3, 342)
30.95 (3, 343)
12.32 (3, 343)
34.16 (3, 324)—
3.92 (3, 324)—
3.32 (3, 324)
14.72 (3, 324)
13.75 (3, 324)—
8.67 (3, 324)
dp < .10.
ep < .05.
fp < .01.
gp < .001.
ab, bc, b, c, and a define homogeneous groups.
614Mateyoke-Scrivner et al.
In addition to drug use, an analysis of criminal activity yielded
significant differences among the four groups. Urban participants who
terminated had more probation violation offenses (M = 2.0, SD = 2.05) than
did any of the other three groups (p < .001) and both urban groups had
more probation violations (M = 2.0, SD = 2.05, M = 1.2, SD = 1.73) than did
the rural groups (M =.2, SD =.54 for rural graduates, and M =.2, SD =.56
for terminated rural participants). In addition, urban graduates had more
miscellaneous criminal offenses (i.e., traffic, prostitution, nonsupport,
miscellaneous, and weapons) (M = 16.8, SD = 12.66) than did any other
group (p < .001). Overall, urban participants had more convictions than did
rural participants, and urban graduates not only had more convictions
(M = 15.7, SD = 11.44) than did rural graduates (M = 8.1, SD = 7.76), but
they also had more convictions than terminated rural participants had
(M = 8.6, SD = 9.51).
Logistic Regression Model
An overall logistic regression model predicting graduation and ter-
mination revealed several predicting variables for drug court termination
(see Table 2). Specifically, for every year increase in a participant’s age,
there was a nearly 5% greater likelihood of graduation from drug court
(b =.05, p < .001). Also, clients employed full-time while in drug court
were 2.9 times more likely to graduate (b =.39, p < .01). Furthermore,
married clients were 57% less likely to graduate (b = ?.85, p < .01).
Results showed that for every year of education, a drug court participant
was 15% more likely to graduate (b =.14, p < .05). Finally, nonwhite drug
court participants were 37% less likely to graduate than were their white
counterparts (b = ?.47, p < .05).
Client patterns of cocaine use in the six months before drug court also
emerged as a significant predictor of drug court graduation. Specifically,
clients who used cocaine more often in the six months before drug court
entry were 14.4% less likely to graduate (b = ?.16, p < .001). Criminal
history variables also were significant predictors for graduation. Clients
more likely to graduate had drug or alcohol charges (7.2% more likely,
b =.07, p < .05) or miscellaneous charges (4.5% more likely, b =.04,
p < .01), while clients less likely to graduate had more probation violations
(30% less likely, b = ?.35, p < .001) and more violent charges (23.5% less
likely, b = ?.27, p < .01).
Separate logistic regression models were conducted for urban and rural
subsamples to explore whether graduation/termination predictors would be
different. For urban participants, those who were not married were 3.4
times more likely to graduate (b = 1.22, p < .05), and those who were
Retention Predictors 615
employed full-time after drug court entry were 3.2 times more likely to
graduate (b =.49, p < .01). For drug use, the more often participants used
cocaine, the less likely they were to graduate (16.5%) (b = ?.19, p < .01).
Like the overall logistic model, participants were 30% less likely to
graduate with every probation violation received (b = ?.36, p < .001) and
5.9% more likely to graduate with each miscellaneous offense (b =.06,
p < .01). For rural participants, however, age was one of the only significant
predictors of graduation or termination. Specifically, for every one year
increase in a participant’s age, graduation was 6% more likely (b =.06,
p < .01). Juvenile incarceration also predicted graduation or termination for
the rural participants with individuals who were incarcerated as a juvenile
being 61% less likely to graduate (b = ?.94, p < .01) than were those not
incarcerated as a juvenile.
Predictorb SEOdds ratio
Overall logistic model
Pattern of cocaine use
Urban logistic model
Pattern of cocaine use
Rural logistic model
Ever locked up before 18
Note: Race: 1 = White, 2 = Nonwhite; Employment: 0 = Unemployed, 1 = Parttime,
2 = Fulltime; Marital Status: 0 = Not Married, 1 = Married; Pattern of cocaine use:
0 = Never, 1 = Only 1–3 times, 2 = About one time per month; 3 = About 2–3 times
per month; 4 = About one time per week; 5 = About 2–6 times per week; 6 = About
one time per day; 7 = About four or more times per day.
616 Mateyoke-Scrivner et al.
To our knowledge, research on drug courts has not compared clients
who complete to clients who do not complete drug court programs across
urban and rural settings. Furthermore, previous treatment retention
comparisons produced contradictory or inconclusive findings. Consequent-
ly, this study focused on comparing and describing differences between
drug court graduates and those who terminated from drug court in one rural
and one urban Kentucky Drug Court.
Graduated vs. Terminated
Overall, when terminated clients were compared with graduated clients,
or those who completed treatment, graduates were older, not married, more
educated, and more likely to be employed full-time. With the exception of
marital status, these findings are supported by previous studies
(7,16,18,20,25–32), which also indicate that clients who completed
treatment were more likely to be older, more educated, and employed
full-time. However, there were no significant findings for gender, race, and
number of children in this study. As expected, this analysis found
terminated drug court clients had more extensive drug use histories than
did graduates. While clients who terminated had more probation violations,
these clients did not have more charges or convictions when compared to
graduates. However, graduates had more drug and alcohol charges, which
may suggest that the drug court programs may be more focused on drugs
and alcohol rather than other criminal involvement and; thus, the drug court
program was more tailored to their needs. Similarly, as expected, drug court
graduates were less likely to be involved in the criminal justice system.
Jacoby (55) suggests that individuals who violate rules and norms in one
area are likely to commit law-breaking activities in other areas (7).
Consequently, those drug court clients who are more involved in criminal
activity at drug court entry are more likely to maintain rule-breaking
behavior and to ultimately be terminated from the drug court program.
Urban vs. Rural
When urban participants were compared to rural participants, rural
participants, as expected, were more likely to be white. Urban participants
had more children and were more likely to be employed full-time. Unlike
previous studies, rural participants reported less education and were not
older than urban participants. However, rural participants reported more
Retention Predictors 617
legal income and less illegal income in the 30 days before drug court. As
noted in other studies, rural participants used cocaine, marijuana, and
multiple substances less frequently than urban participants did. Perhaps the
difference in drug use between rural and urban areas could be partially
explained by differences in availability of certain substances in rural areas
when compared to urban areas. Cocaine, for example, is more prevalent in
urban areas because it is more accessible.
No differences in mental health problems were noted, however, when
urban and rural drug courts were compared. It may be possible that mental
health problems are more associated with poverty in urban rather than rural
areas. Rural clients in this study reported slightly more income than did
urban participants. There were no differences in mental health measures.
Also, it has been thought that residing in a rural area could be a protective
factor from mental health problems because rural areas are generally closer-
knit communities, with greater social support systems (56). In addition,
rural drug court participants had fewer charges and convictions than did
urban participants, which may be explained by residents being less likely to
report crimes in rural areas due to a desire to keep things within the
confines of a rural community (45).
Graduation Status by Urban/Rural
When the four groups were compared, rural drug court graduates were
older than were terminated rural clients and both rural and urban graduate
groups had more years of education and more full-time employment than
either terminated group. These findings are consistent with findings from
other studies (28–32). Both urban graduates and rural graduates were more
likely to be nonmarried when compared to the terminated clients. This
finding contradicts previous ideas that having a spouse or significant other
helps to provide the social support needed to complete treatment (57). While
this finding does not support previous studies, perhaps clients who were not
married had fewer family obligations and, therefore, more time to devote to
the drug court program. Another possible explanation is that because the
current study had a higher percentage of nonmarried clients participating
(82.1% vs. 17.6%), nonmarried clients may be more likely to graduate
because a greater number of nonmarried clients participated in this study.
When drug use was compared, there were few differences. Findings
show that urban graduates used cocaine and multiple substances less
frequently and for shorter periods of time than did urban clients who
terminated. In addition, rural clients who terminated from drug court
reported fewer years of cocaine use than either group of graduates did.
Findings from previous studies (35,36) indicate that anxiety and
depression are common among substance abusers. Consequently, drug court
618 Mateyoke-Scrivner et al.
graduates in this study were expected to report fewer mental health
problems than were those who terminated. However, this study found no
significant differences between these groups on anxiety and only slightly
more depression for those who terminated.
The criminal history and treatment retention literature indicates that
early experiences with the criminal justice system, as well as more
extensive criminal histories are associated with treatment drop-out (27,41).
This was supported by the current study which found that juvenile
incarceration was related to termination. In addition, both terminated groups
reported juvenile incarceration and violent offenses more frequently than
did graduates. However, for most offenses, both urban groups had more
probation violations, drug and alcohol offenses, miscellaneous criminal
offenses, total charges, and total convictions. As is consistent with the
existing literature (24,43), this study found that both urban and rural
terminated groups reported more violent offenses than did graduates, again
potentially associated with previous rule-breaking behaviors that continued
into drug court (45).
Like other studies on treatment retention (25), the current study found
that being white is a predictor of drug court retention. While race/ethnicity
might be associated with termination or completion from drug court, it is
possible that treatment success might be more related to other variables
such as less education or less income. In contrast with other studies, this
study found no significant differences in treatment retention between males
and females in either rural or urban drug court settings.
Findings from this study support education and employment findings
from previous studies. Specifically, participants with more education and
those who were employed full-time after entering drug court were more
likely to complete treatment. This study also supports previous research
(17,24,25,27), related to age as a predictor of treatment completion. For the
overall and the rural regression models, older participants were more likely
to graduate from the drug court program. Interestingly, even though rural
participants were about the same age as urban participants, age was a
predictor of drug court completion for rural drug court participants.
Overall, this study is consistent with the literature on treatment
retention since participants with more severe drug use histories were less
likely to complete treatment (7,58). Cocaine use was a significant predictor
for termination across each group. Specifically, the more cocaine was used
and the longer cocaine was used was related to a decreased likelihood that a
drug court client would graduate from the program. For urban clients,
however, the pattern of cocaine use was a significant predictor of
noncompletion (e.g., the more cocaine a client used, the less likely the
client was to graduate). However, the years of cocaine use were not related
to completion. Perhaps the intensity of cocaine use, rather than the duration
of the addition, is a predictor of treatment completion. In fact, research has
shown that cocaine, unlike other drugs, creates a more intense addiction and
can become addictive in a few weeks or months as opposed to years with
other drugs. This more intense addiction may explain why cocaine users
were less likely to complete treatment (59). Unlike cocaine use, while
multiple drug use differed for each group, it was not a predictor of
This study also found no significant differences between completers
and noncompleters on Brief Symptoms Index (BSI) mental health variables
unlike findings from other drug treatment studies. Again, there could be
something unique about drug court clients that is different from other drug
treatment clients. Because the BSI measures mental health within the last
seven days, it may be possible that the time focus by this instrument did not
cover an adequate period of time. In addition, BSI measures are self-
reported and not clinically diagnosed; therefore, a client’s mental health
could be misrepresented. Also, according to the literature (36), rural
participants are more likely to report psychological problems; however,
there were no differences for rural clients since all four groups reported
similar responses to mental health problems.
There was also a significant relationship between criminal activity and
treatment retention. Specifically, drug court termination was related to more
probation offenses, violent offenses, and other criminal offenses. Interest-
ingly, prior drug or alcohol offenses increased a participant’s odds of
graduating. For the urban drug court, more probation offenses were a
predictor of drug court termination, while each other criminal offense made
graduation more likely. Like other studies (27,41) rural participants who
were incarcerated as a juvenile were less likely to graduate from drug court.
While several studies (7,42) suggest that the number of charges and the
number of convictions are related to treatment dropout, this study did not
support the number of total charges and convictions as predictors of
completing drug court. Perhaps, because previous studies involved other
treatment programs, there may be some unique quality about drug court
clients that warrants further research.
There are limitations to this study. Participants were chosen after
eligibility for drug court was determined. Thus, participants were not
620 Mateyoke-Scrivner et al.
randomly selected. Only drug courts from one state were chosen; therefore,
findings may not be generalizable to other drug courts within the state or in
other states. In addition, data were self-reported, which produces recall
limitations and misrepresentation. While participants volunteered and
consented to participate in the study, it is not known how truthful
participants were about their self-reported behaviors.
Despite these limitations, this study adds to the literature on drug
abuse treatment and drug court treatment specifically by identifying
differences between treatment predictors for urban and rural drug court
participants. By understanding treatment retention, drug court personnel
and others could use the overall study findings as part of assessing
potential participants, how a client’s background might affect treatment
outcomes, and to tailor services to the needs of each client, such as
targeting mental health services and family services for specific
individuals. Group sessions could target the special needs of drug court
clients and community resources could be focused to target these needs.
Focusing on factors related to treatment retention could not only benefit
drug court clients by providing a more individualistic treatment program,
but it could also boost the number of graduates for the drug court program,
possibly creating more funding and program acceptance. In addition, drug
court programs could include lifestyle changes in addition to the current
focus on no drug and alcohol use. For example, a client in drug court for
trafficking may be more addicted to the money and lavish lifestyle that
comes with the act rather than the drug use itself. It is also important for
drug court programs to note the differences in needs between rural drug
court clients and the urban drug court clients. For example, by
understanding that being employed full-time, for an urban drug court
client, is a factor in treatment retention, drug court administrators could
make job retention a higher priority and devote more resources to finding
jobs for clients. Future research on graduation/termination predictors for
drug court clients could include motivation and social support as variables
in order to better understand treatment outcomes. In addition, contextual
factors, such as rural and urban environments, should be considered in
This study is supported by a grant from the National Institute on
Drug Abuse, Grant #DA 13076. Opinions expressed in this paper are
those of the authors and do not represent the view the National Institute
on Drug Abuse.
1. Wilson DJ. Drug Use, Testing, and Treatment in Jails, NCJ Publication
No. 9999. Washington: U.S. Department of Justice, D.C., 2000.
Sherin KM, Mahoney B. Treatment Drug Courts: Integrating Substance
Abuse Treatment with Legal Case Processing. Publication No. (SMA)
96-3113). U.S. Department of Health and Human Services, 1996.
Belenko S. Research on Drug Courts: A Critical Review. The National
Center on Addiction and Substance Abuse at Columbia University,
Defining Drug Courts: The Key Components. Department of Justice,
Office of Justice Programs, Drug Court Programs Office: Washington,
Staton M, Mateyoke A, Leukefeld CG, Cole J, Hopper H, Logan TK,
Minton L. Employment issues among drug court participants. J Offender
Rehabil 2001; 33(4):73–85.
Drug Court Activity Update: Summary Information on All Programs
and Detailed Information on Adult Drug Courts. Drug Court
Clearinghouse and Technical Assistance Project, U.S. Department of
Justice: Washington, DC, 2001.
Lang MA, Belenko S. Predicting retention in a residential drug
treatment alternative to prison program. J Subst Abuse Treat 2000;
Prendergast ML, Anglin MD. Treatment for drug-abusing offenders
under community supervision. Fed Probat 1995; 59(4):66–76.
Roberts AC, Nishimoto RH. Predicting treatment retention of
women dependent on cocaine. Am J Drug Alcohol Abuse 1996;
Simpson DD, Joe GW, Rowan-Szal GA. Drug abuse treatment retention
and process effects on follow-up outcomes. Drug Alcohol Depend
Bell DC, Richard AJ, Feltz LC. Mediators of drug treatment outcomes.
Addict Behav 1996; 21(5):597–613.
Bachman JG, Wadsworth K, O’Malley PM, Johnston LD, Schulenberg
J. Smoking, Drinking, and Drug Use in Young Adulthood. Mahwah,
NJ: Earlbaum, 1997.
Kandel DB, Yamaguchi K. Job mobility and drug use: an event history
analysis. Am J Sociol 1987; 92:836–878.
Sampson RJ, Laub JH. Crime in the Making. Cambridge: Harvard
Cresswell LS. Minority and non-minority perceptions of drug court
program severity and effectiveness. J Drug Issues 2001; 31(1):259–293.
622Mateyoke-Scrivner et al.
16. Sayre SL, Schmitz JM, Stotts AL, Averill PM, Rhoades HM,
Grabowski JJ. Determining predictors of attrition in an outpatient
substance abuse program. Am J Drug Alcohol Abuse 2002; 28(1):55–
Study finds predictors of treatment completion sex-specific. Alcohol
Drug Abuse Week 2000; 12(41):3.
Allison M, Hubbard RL. Drug abuse treatment process: a review of the
literature. Int J Addict 1985; 20:1321–1345.
Hiller ML, Knight K, Broome KM, Simpson DD. Legal pressure and
treatment retention in a national sample of long-term residential
programs. Crim Justice Behav 1998; 25(4):463–481.
McLellan AT. Patient characteristics associated with outcome. In:
Research on the Treatment of Narcotic Addiction: State of the Art,
DHHS Publication No. LADM 83-1281; Washington D.C.: Government
Printing Office, 1983.
Chou CP, Hser YI, Anglin MD. Interaction effects of client and
treatment program characteristics on retention: an exploratory analysis
using hierarchical liner models. Subst Use Misuse 1998; 33:2281–
Simpson DD, Joe GW, Broome KM, Hiller ML, Knight K, Rowan-Szal
GA. Program diversity and treatment retention rates in drug abuse
treatment outcome study (DATOS). Psychol Addict Behav 1997;
Melnick G, De Leon G, Hawke J, Jainchill N, Kressel D. Motivation
and readiness for therapeutic community treatment among adolescents
and adult substance abusers. Am J Drug Alcohol Abuse 1997;
Saum CA. Violent offenders in drug court. J Drug Issues 2001;
Sechrest DK. Determinants of graduation from a day treatment drug
court in California: a preliminary study. J Drug Issues 2001; 31(1):129–
Joe GW, Simpson DD. Retention in treatment of drug users: 1971–
1972 DARP admissions. Am J Drug Alcohol Abuse 1975; 2:63–71.
Sung HE, Belenko S, Feng L. Treatment compliance in the trajectory of
treatment progress among offenders. J Subst Abuse Treat 2001;
Comerford AW. Work dysfunction and addiction: common roots.
J Subst Abuse Treat 1999; 16:247–253.
Leukefeld C.G., Smiley H., Staton M., Mateyoke-Scrivner A.
Employment, Employment Related Problems, and Drug Use at Drug
Court Entry. 2003. Submitted to Subst. Use Misuse.
Retention Predictors 623
30. Vaillant GE. What can long-term follow-up teach us about relapse and
prevention of relapse in addiction. Br J Addict 1988; 83:1147–1157.
Zanis DA, Metzger DS, McLellan A.T. Factors associated with
employment among methadone patients. J Subst Abuse Treat 1995;
Wolkstein E, Bausch R, Weber G. Work as a Critical Component of
Recovery, 2000. http://www.med.wright.edu/citar/sardi/workandrecovery.
Winick C. Maturing out of narcotic addiction. B Narcotics 1962;
Leukefeld CG, Clayton RR, Myers JA. Rural drug and alcohol
treatment. Drugs Soc 1992; 7(1 and 2):95–116.
Galanter M, Castaneda R, Ferma J. Substance abuse among general
psychiatric patients: place of presentation, diagnosis, and treatment. Am
J Drug Alcohol Abuse 1988; 14(2):211–235.
Robertson EB, Donnermeyer JF. Illegal drug use among rural adults:
mental health consequences and treatment utilization. Am J Drug
Alcohol Abuse 1997; 23(3):467–485.
Schmitz JM, Bordnick P, Kearney M, Fuller SM, Breckenridge JK.
Treatment outcome of cocaine-alcohol dependent patients. Drug
Alcohol Depend 1997; 47:55–61.
Brown BS, Voskuhl TC, Lehman PE. Comparison of drug abuse clients
in urban and rural settings. Am J Drug Alcohol Abuse 1977; 4(4):445–
Blazer D, George LK, Landerman R, Pennybacker M, Melville ML,
Woodbury M, Manton KG, Jordan K, Locke B. Psychiatric disorders: a
rural/urban comparison. Arch Gen Psychiatry 1985; 42(7):651–656.
Logan TK, Schenck JE, Leukefeld CG, Meyers JA, Allen S. Rural
attitudes, opinions, and drug use. Subst Use Misuse 1999; 3&4(45):
Gainey RR, Wells EA, Hawkins JW, Catalano RF. Predicting treatment
retention among cocaine users. Int J Addict 1993; 28(6):487–505.
De Leon G, Melnick G, Kressel D. Motivation and readiness for
therapeutic community treatment among cocaine and other drug
abusers. Am J Drug Alcohol Abuse 1997; 23(2):169–190.
Broome KM, Flynn PM, Simpson DD. Psychiatric comorbidity
measures as predictors of retention in drug abuse treatment programs.
Health Serv Res 1999; 34(3):791.
Van Stelle KR, Mauser E. Recidivism to the criminal justice system of
substance-abusing offenders diverted into treatment. Crime Delinq
624Mateyoke-Scrivner et al.
45.Weisheit RA, Wells LE. Rural crime and justice: implications for
theory and research. Crime Delinq 1996; 42(3):379–397.
Appalachia: Twenty Years of Progress. Appalachian Regional Com-
mission: Washington, DC, 1985.
Duncan CM, Tickamyer AR. Poverty research and policy for rural
America. Am Sociol 1988; 19(3):243–258.
Brownsberger WN. Prevalence of frequent cocaine use in urban poverty
areas. Contemp Drug Probl 1997; 24:349–371.
Wagenfeld MO. Mental health and rural America: a decade review.
J Rural Health 1990; 6:1–27.
Kentucky Administrative Office of the Courts. 2002. Retrieved Decem-
ber 2002. http://www.kycourts.net/AOC/drugcourt/AOC_DC_Sites.
Logan TK, Hoyt W, Leukefeld CG. Kentucky Drug Court Outcome
Evaluation: Behavior, Costs, and Avoided Costs to Society. Lexington:
University of Kentucky, Center on Drug and Alcohol Research, 2001.
McLellan AT, Luborsky L, O’Brien CP, Woody GE. An improved
diagnostic instrument for substance abuse patients: the addiction
severity index. J Nerv Ment Dis 1980; 168:26–33.
Kosten TR, Rounsaville BJ, Kleber HA. Parental alcoholism in opioid
addicts. J Nerv Ment Dis 1987; 173:461–468.
Derogatis IR. The SCL-90: Administration, Scoring, and Procedures
Manual-II. Clinical Psychometric Research: Towson, MD, 1983.
Classics of Criminology. Jacoby JE, ed. Prospect Heights, IL:
Waveland Press, 1988.
Leukefeld CG, Narevic E, Hiller ML, Staton M, Logan TK, Gillespie
W, Webster JM, Garrity TF, Purvis R. Alcohol and drug use among
rural and urban incarcerated substance abusers. Int J Offender Ther
Stark MJ. Dropping out of substance abuse treatment: a clinically
oriented review. Clin Phys Rev 1992; 2:93–116.
Miller JM. Considering the need for empirically grounded drug court
screening mechanisms. J Drug Issues 2001; 31:91–107.
Lowinson JH, Ruiz P, Millman RB, Langrod JG. Substance Abuse: A
Comprehensive Textbook. 3rd ed. Baltimore: Williams & Waverly,
Retention Predictors 625
Page 22 Download full-text