image-removebg-preview

Graduated risk: Individual- and program-level predictors of drug court completion

  • Home
  • Research
  • Graduated risk: Individual- and program-level predictors of drug court completion

Participation in drug court is associated with reduced recidivism and drug use for individuals in the criminal justice system. However, many drug court participants do not complete their respective programs. The authors of this study examined which characteristics of drug court participants and programs predicted higher rates of drug court completion.

Stay on the frontiers of
recovery science
with the free, monthly
Recovery Bulletin

l

WHAT PROBLEM DOES THIS STUDY ADDRESS?

Substance use disorders are prevalent among incarcerated people in the United States. Drug courts offer an alternative to incarceration and an opportunity to receive substance use disorder treatment for certain individuals within the criminal justice system. Drug court participation is associated with reduced rates of recidivism (i.e., re-arrest) and drug use; however, a large subset (30-50%) of drug court participants do not complete their respective programs – also known as “graduation”.

Determining the factors that influence drug court completion is a complicated endeavor. Most of the existing research has focused on which characteristics of drug-court participants (i.e., individual-level variables) predict graduation from these programs. Generally, participants who are older, female, White (vs. non-White), employed, and more educated are more likely to graduate from drug court programs. On the other hand, individuals with more serious criminal histories and more severe substance use and mental health difficulties are less likely to graduate. That said, individuals with more severe needs or complex histories are less likely to be admitted to drug courts in the first place, potentially skewing these results. Additionally, there has been less research in samples with greater racial/ethnic diversity.

Research on how programmatic factors—characteristics of the programs, including requirements, treatment received, etc.— can also help inform strategies to improve graduation rates. A study combining results from 76 studies of drug courts found that, in general, more requirements were related to an increase in program effectiveness, whereas certain requirements (sanctions, fines) were related to a decrease in effectiveness. However, there has been less focus on treatment received during these programs. Focusing on program-level factors is important because these factors can be changed and tailored to individuals’ needs, whereas individual-level characteristics are not necessarily as amenable to change.

In this study, researchers examined both individual and program-level factors that may influence rates of graduation to inform drug court policies and practices.


HOW WAS THIS STUDY CONDUCTED?

This study involved naturalistic follow-up and tracking of graduation outcomes among 247 participations from a large felony-level drug court in a southeastern U.S. city. The drug court had two tracks: 1) “Track 1” was for higher-risk participants (i.e., severe SUD) and involved five treatment phases over 24 months; 2) “Track 2” was for moderate-to-lower risk participants (i.e., those with stable employment, stable housing, fewer felony convictions) and included four treatment phases over 21 months. Participants in both tracks had to attend all court programming and received sanctions for infractions including positive drug screens, violent/destructive behavior, or re-arrest. Sanctions included treatment response (i.e., completing additional homework and/or increased intensity of treatment sessions), community service, jail, behavioral contracts, 30-day step-up (i.e., participants have 30 days to meet agreed-upon terms or are dismissed), and monetary fines and were applied using a matrix with consideration of contextual factors.

The data used in this study came from drug court records as well as arrest data from state records. Drug court records included data on days in program, type and attendance at treatment sessions, and number/type of sanctions issued. All participants completed two assessments upon admission to drug court, including an assessment of various indices of level of risk/need (e.g., criminal history, education/employment history, emotional/personal risk, and other factors) and an assessment of substance use history. Criteria for graduation from drug court included completing all program phases, having no positive drug screens for 6 months prior to graduation, completing an aftercare plan, and completing the final assessment (described earlier) at drug court exit.

The study’s primary purpose was to examine the extent to which individual-level and program-level factors (and their combination) predicted graduation from drug court. To accomplish this, researchers first examined which factors were individually related to graduation. Then they tested all of these factors at once, so that each variable’s unique effect could be examined while accounting for other variables initially related to graduation. This approach helps determine the independent effect of each variable, increasing confidence that the factor is causally related to drug court graduation.

The individual-level factors included variables related to individual substance use history and mental health, and the program-level factors included variables capturing amount and type of treatment received in addition to program requirements. The drug court studied was a large, felony-level drug court in that it included a diverse sample of moderate- and high-risk individuals, allowing the authors to examine a range of risk and participant characteristics.

The final sample included data from 247 drug court participants. Of the full sample, 77.7% were male, 61.8% were Black, 53.8% graduated high school, and 38.4% were employed; the average age was 38.9 years. 55.4% of participants reported cocaine as their drug of choice, 81.9% endorsed using a “criminogenic” drug of choice (i.e., crack cocaine/cocaine, heroin/other opiates, and methamphetamines/other amphetamines), and 50.9% reported using multiple substances. The average number of prior arrests was 17.2 total arrests and 9.9 felony arrests.


WHAT DID THIS STUDY FIND?

Less than half of participants graduated from drug court.

A total of 46% of participants graduated from drug court. Track 1 (i.e., higher risk track) participants were less likely to graduate from drug court than participants in Track 2 (38% vs. 55%).

More program factors were related to graduation than individual factors.

Only one individual-level factor was associated with graduation rates in the final combined analyses. Higher levels of emotional/personal risk (i.e., an index containing items assessing mental health, emotional management, and antisocial personality features) was associated with lower odds of graduation. Having a high score on this measure was associated with having a 44% decrease in the odds of graduation from drug court.

Multiple program-level factors predicted odds of graduation. Specifically, having a greater number of days in program and individual counseling sessions predicted higher odds of graduation, whereas having a greater number of sanctions decreased odds of graduation. Controlling for the number of days in the program, having more individual sessions was associated with a 20% increase in the odds of graduating. A greater number of jail and monetary sanctions predicted lower odds of graduation, decreasing the odds of graduation by 65% and 71%, respectively.


WHAT ARE THE IMPLICATIONS OF THE STUDY FINDINGS?

This study examined the effects of individual and programmatic factors on graduation from a large, urban drug court. Less than half of the drug court participants in the current study graduated from drug court, which is consistent with prior estimates of drug court graduation rates. In analyses examining both individual- and program-level predictors of drug court graduation, only one individual-level variable predicted graduation rates, whereas several program-level variables related to graduation.

Specifically, the only individual-level predictor of graduation, having higher levels of emotional/personal risk at baseline—which represents severity of current/past mental health conditions including current antisocial personality disorder—was related to lower likelihood of drug court graduation. Although personal mental health history is an individual variable, it does highlight an area that could be addressed when designing drug court programming. For instance, a pilot study of a drug court program that included a set of interventions addressing co-occurring mental health conditions (MISSION-CJ) showed promising outcomes. Participants who received this intervention had high levels of continued engagement in treatment, as well as reduced rates of recidivism (measured as nights in jail) and decreased co-occurring psychiatric and trauma symptoms. More rigorously designed studies comparing this type of intervention to traditional programming will be helpful in determining whether an increased focus on co-occurring psychiatric disorders can improve outcomes and graduation rates among drug court participants.

The focus on how programmatic features can influence graduation outcomes is important, as these features can be changed, whereas aspects of individuals’ backgrounds cannot. Program features that were associated with graduation in this study include those related to increased treatment – i.e., greater number of days in program and greater number of individual treatment sessions. By contrast, a greater number of punitive actions, particularly more sanctions (particularly jail and monetary), were associated with a decreased likelihood of graduation. However, this study used a design that was correlational, so it is unclear if these program features actually caused poorer outcomes or reflected individual differences. For instance, those who attended more treatment sessions may have just been more motivated and adherent to the programming itself, whereas those at higher individual risk (i.e., due to more severe SUD, co-occurring mental health difficulties, etc.) also may have been more likely to receive sanctions due to their behavior/non-adherence. While researchers did their best to address some of these alternative explanations for the findings with the study design, there were many factors for which they could not account given the data at their disposal. More research, ideally comparing drug courts with different intervention/sanction structures, is needed to elucidate the degree to which these program features impact graduation in a causal way.

The detrimental effect of sanctions has been found in several other studies. As noted, receiving sanctions can be indicative of other risk factors and can signify that a participant is nearing termination from drug court. Some researchers have started to study ways that sanctions can be implemented so that they are more likely to promote good outcomes. Specifically, sanctions that are “swift, certain, and fair” are associated with lower rates of substance use; however, their relationship with crime recidivism is less conclusive. Additional research examining ways that sanctions can be employed to promote positive behavior change is needed.


  1. This study’s design was observational, thus relationships between individual- and program-level variables and graduation rates are correlational in nature. Further research is needed to clarify the degree to which these variables directly impact graduation rates and whether there are relevant interactions between variables and/or additional intervening variables that could better explain these associations.
  2. The data included in this study was limited to what was available in records kept by the drug court, which in turn limited the amount of detail regarding type/quality of treatment, prior substance use history, prior treatment history, or time in recovery housing.
  3. The data in this study did not include information regarding the reasons why participants dropped out of drug court, which could better explain factors contributing to drop-out (e.g., mental health, motivational, other psychosocial factors, etc.).
  4. The sample only included individuals from one drug court. Although this drug court included a diverse set of participants representing a range of risk, findings cannot necessarily be generalized to other parts of the country.

BOTTOM LINE

This study examined the effect of individual and programmatic factors on a large, urban drug court. Less than half of participants in this drug court completed the intended duration of the treatment program and met its goals. When multiple individual-level and program-level predictors were examined at once, researchers found that having higher levels of co-occurring mental health/personality difficulties was associated with lower likelihood of graduating from drug court. At the program-level, a higher number of days in the program and more individual treatment sessions were associated with a greater likelihood of graduation, whereas a higher number of sanctions (particularly jail or monetary sanctions) were associated with lower graduation rates. Additional research is needed to clarify whether adding more individual counseling sessions, changing aspects of drug court programming to address mental health needs, and/or designing sanctions to be more effectively delivered, can improve outcomes.


  • For individuals and families seeking recovery: For individuals involved in the criminal justice system, drug court can be a viable option to avoid incarceration while also participating in programming to address a substance use disorder and reduce substance use. Having co-occurring mental health difficulties can reduce the likelihood of graduating from drug court; thus, supplemental mental health treatment or seeking additional individual intervention within a drug court program may better promote success. 
  • For treatment professionals and treatment systems: Participants in drug court are likely to benefit from continued support and interventions for co-occurring mental health conditions from treatment professionals. In addition to addressing baseline mental health needs, providers working with drug court participants may also assist patients with modifying behaviors that may interfere with successful engagement with drug court programming and ultimately graduation (e.g., motivation, other barriers to participation in treatment, etc.).
  • For scientists: These research findings suggest several possible avenues for further drug court research. Most existing research has been correlational or limited to pilot studies. More rigorously designed studies—randomized controlled trials or quasi-experimental studies with appropriate comparison groups and rigorous designs (e.g., propensity score matched comparison group)—are needed to determine if additional intervention (e.g., more counseling sessions; interventions designed to address co-occurring psychiatric disorders in addition to substance use disorder) or changes to drug court program structure (e.g., changes in application or delivery of sanctions) directly impacts graduation rates. Given regional differences in drug court programming, future research should include multiple drug courts from different regions to gauge generalizability of findings. 
  • For policy makers: More funding for drug courts and research into drug court programming can help clarify aspects of programming that are effective as well as aspects of programming that can be changed/altered to improve outcomes. Given that drug courts can improve rates of substance use and recidivism, improving graduation rates and successful engagement in these programs is likely to have broader benefits for the criminal justice system and society at large. 

CITATIONS

Randall-Kosich, O., Whitaker, D. J., Guastaferro, W. P., & Rivers, D. (2022). Predicting drug court graduation: Examining the role of individual and programmatic characteristics. Journal of Substance Abuse Treatment, 135, 108654. doi: 10.1016/j.jsat.2021.108654


Stay on the Frontiers of
recovery science
with the free, monthly
Recovery Bulletin

l

WHAT PROBLEM DOES THIS STUDY ADDRESS?

Substance use disorders are prevalent among incarcerated people in the United States. Drug courts offer an alternative to incarceration and an opportunity to receive substance use disorder treatment for certain individuals within the criminal justice system. Drug court participation is associated with reduced rates of recidivism (i.e., re-arrest) and drug use; however, a large subset (30-50%) of drug court participants do not complete their respective programs – also known as “graduation”.

Determining the factors that influence drug court completion is a complicated endeavor. Most of the existing research has focused on which characteristics of drug-court participants (i.e., individual-level variables) predict graduation from these programs. Generally, participants who are older, female, White (vs. non-White), employed, and more educated are more likely to graduate from drug court programs. On the other hand, individuals with more serious criminal histories and more severe substance use and mental health difficulties are less likely to graduate. That said, individuals with more severe needs or complex histories are less likely to be admitted to drug courts in the first place, potentially skewing these results. Additionally, there has been less research in samples with greater racial/ethnic diversity.

Research on how programmatic factors—characteristics of the programs, including requirements, treatment received, etc.— can also help inform strategies to improve graduation rates. A study combining results from 76 studies of drug courts found that, in general, more requirements were related to an increase in program effectiveness, whereas certain requirements (sanctions, fines) were related to a decrease in effectiveness. However, there has been less focus on treatment received during these programs. Focusing on program-level factors is important because these factors can be changed and tailored to individuals’ needs, whereas individual-level characteristics are not necessarily as amenable to change.

In this study, researchers examined both individual and program-level factors that may influence rates of graduation to inform drug court policies and practices.


HOW WAS THIS STUDY CONDUCTED?

This study involved naturalistic follow-up and tracking of graduation outcomes among 247 participations from a large felony-level drug court in a southeastern U.S. city. The drug court had two tracks: 1) “Track 1” was for higher-risk participants (i.e., severe SUD) and involved five treatment phases over 24 months; 2) “Track 2” was for moderate-to-lower risk participants (i.e., those with stable employment, stable housing, fewer felony convictions) and included four treatment phases over 21 months. Participants in both tracks had to attend all court programming and received sanctions for infractions including positive drug screens, violent/destructive behavior, or re-arrest. Sanctions included treatment response (i.e., completing additional homework and/or increased intensity of treatment sessions), community service, jail, behavioral contracts, 30-day step-up (i.e., participants have 30 days to meet agreed-upon terms or are dismissed), and monetary fines and were applied using a matrix with consideration of contextual factors.

The data used in this study came from drug court records as well as arrest data from state records. Drug court records included data on days in program, type and attendance at treatment sessions, and number/type of sanctions issued. All participants completed two assessments upon admission to drug court, including an assessment of various indices of level of risk/need (e.g., criminal history, education/employment history, emotional/personal risk, and other factors) and an assessment of substance use history. Criteria for graduation from drug court included completing all program phases, having no positive drug screens for 6 months prior to graduation, completing an aftercare plan, and completing the final assessment (described earlier) at drug court exit.

The study’s primary purpose was to examine the extent to which individual-level and program-level factors (and their combination) predicted graduation from drug court. To accomplish this, researchers first examined which factors were individually related to graduation. Then they tested all of these factors at once, so that each variable’s unique effect could be examined while accounting for other variables initially related to graduation. This approach helps determine the independent effect of each variable, increasing confidence that the factor is causally related to drug court graduation.

The individual-level factors included variables related to individual substance use history and mental health, and the program-level factors included variables capturing amount and type of treatment received in addition to program requirements. The drug court studied was a large, felony-level drug court in that it included a diverse sample of moderate- and high-risk individuals, allowing the authors to examine a range of risk and participant characteristics.

The final sample included data from 247 drug court participants. Of the full sample, 77.7% were male, 61.8% were Black, 53.8% graduated high school, and 38.4% were employed; the average age was 38.9 years. 55.4% of participants reported cocaine as their drug of choice, 81.9% endorsed using a “criminogenic” drug of choice (i.e., crack cocaine/cocaine, heroin/other opiates, and methamphetamines/other amphetamines), and 50.9% reported using multiple substances. The average number of prior arrests was 17.2 total arrests and 9.9 felony arrests.


WHAT DID THIS STUDY FIND?

Less than half of participants graduated from drug court.

A total of 46% of participants graduated from drug court. Track 1 (i.e., higher risk track) participants were less likely to graduate from drug court than participants in Track 2 (38% vs. 55%).

More program factors were related to graduation than individual factors.

Only one individual-level factor was associated with graduation rates in the final combined analyses. Higher levels of emotional/personal risk (i.e., an index containing items assessing mental health, emotional management, and antisocial personality features) was associated with lower odds of graduation. Having a high score on this measure was associated with having a 44% decrease in the odds of graduation from drug court.

Multiple program-level factors predicted odds of graduation. Specifically, having a greater number of days in program and individual counseling sessions predicted higher odds of graduation, whereas having a greater number of sanctions decreased odds of graduation. Controlling for the number of days in the program, having more individual sessions was associated with a 20% increase in the odds of graduating. A greater number of jail and monetary sanctions predicted lower odds of graduation, decreasing the odds of graduation by 65% and 71%, respectively.


WHAT ARE THE IMPLICATIONS OF THE STUDY FINDINGS?

This study examined the effects of individual and programmatic factors on graduation from a large, urban drug court. Less than half of the drug court participants in the current study graduated from drug court, which is consistent with prior estimates of drug court graduation rates. In analyses examining both individual- and program-level predictors of drug court graduation, only one individual-level variable predicted graduation rates, whereas several program-level variables related to graduation.

Specifically, the only individual-level predictor of graduation, having higher levels of emotional/personal risk at baseline—which represents severity of current/past mental health conditions including current antisocial personality disorder—was related to lower likelihood of drug court graduation. Although personal mental health history is an individual variable, it does highlight an area that could be addressed when designing drug court programming. For instance, a pilot study of a drug court program that included a set of interventions addressing co-occurring mental health conditions (MISSION-CJ) showed promising outcomes. Participants who received this intervention had high levels of continued engagement in treatment, as well as reduced rates of recidivism (measured as nights in jail) and decreased co-occurring psychiatric and trauma symptoms. More rigorously designed studies comparing this type of intervention to traditional programming will be helpful in determining whether an increased focus on co-occurring psychiatric disorders can improve outcomes and graduation rates among drug court participants.

The focus on how programmatic features can influence graduation outcomes is important, as these features can be changed, whereas aspects of individuals’ backgrounds cannot. Program features that were associated with graduation in this study include those related to increased treatment – i.e., greater number of days in program and greater number of individual treatment sessions. By contrast, a greater number of punitive actions, particularly more sanctions (particularly jail and monetary), were associated with a decreased likelihood of graduation. However, this study used a design that was correlational, so it is unclear if these program features actually caused poorer outcomes or reflected individual differences. For instance, those who attended more treatment sessions may have just been more motivated and adherent to the programming itself, whereas those at higher individual risk (i.e., due to more severe SUD, co-occurring mental health difficulties, etc.) also may have been more likely to receive sanctions due to their behavior/non-adherence. While researchers did their best to address some of these alternative explanations for the findings with the study design, there were many factors for which they could not account given the data at their disposal. More research, ideally comparing drug courts with different intervention/sanction structures, is needed to elucidate the degree to which these program features impact graduation in a causal way.

The detrimental effect of sanctions has been found in several other studies. As noted, receiving sanctions can be indicative of other risk factors and can signify that a participant is nearing termination from drug court. Some researchers have started to study ways that sanctions can be implemented so that they are more likely to promote good outcomes. Specifically, sanctions that are “swift, certain, and fair” are associated with lower rates of substance use; however, their relationship with crime recidivism is less conclusive. Additional research examining ways that sanctions can be employed to promote positive behavior change is needed.


  1. This study’s design was observational, thus relationships between individual- and program-level variables and graduation rates are correlational in nature. Further research is needed to clarify the degree to which these variables directly impact graduation rates and whether there are relevant interactions between variables and/or additional intervening variables that could better explain these associations.
  2. The data included in this study was limited to what was available in records kept by the drug court, which in turn limited the amount of detail regarding type/quality of treatment, prior substance use history, prior treatment history, or time in recovery housing.
  3. The data in this study did not include information regarding the reasons why participants dropped out of drug court, which could better explain factors contributing to drop-out (e.g., mental health, motivational, other psychosocial factors, etc.).
  4. The sample only included individuals from one drug court. Although this drug court included a diverse set of participants representing a range of risk, findings cannot necessarily be generalized to other parts of the country.

BOTTOM LINE

This study examined the effect of individual and programmatic factors on a large, urban drug court. Less than half of participants in this drug court completed the intended duration of the treatment program and met its goals. When multiple individual-level and program-level predictors were examined at once, researchers found that having higher levels of co-occurring mental health/personality difficulties was associated with lower likelihood of graduating from drug court. At the program-level, a higher number of days in the program and more individual treatment sessions were associated with a greater likelihood of graduation, whereas a higher number of sanctions (particularly jail or monetary sanctions) were associated with lower graduation rates. Additional research is needed to clarify whether adding more individual counseling sessions, changing aspects of drug court programming to address mental health needs, and/or designing sanctions to be more effectively delivered, can improve outcomes.


  • For individuals and families seeking recovery: For individuals involved in the criminal justice system, drug court can be a viable option to avoid incarceration while also participating in programming to address a substance use disorder and reduce substance use. Having co-occurring mental health difficulties can reduce the likelihood of graduating from drug court; thus, supplemental mental health treatment or seeking additional individual intervention within a drug court program may better promote success. 
  • For treatment professionals and treatment systems: Participants in drug court are likely to benefit from continued support and interventions for co-occurring mental health conditions from treatment professionals. In addition to addressing baseline mental health needs, providers working with drug court participants may also assist patients with modifying behaviors that may interfere with successful engagement with drug court programming and ultimately graduation (e.g., motivation, other barriers to participation in treatment, etc.).
  • For scientists: These research findings suggest several possible avenues for further drug court research. Most existing research has been correlational or limited to pilot studies. More rigorously designed studies—randomized controlled trials or quasi-experimental studies with appropriate comparison groups and rigorous designs (e.g., propensity score matched comparison group)—are needed to determine if additional intervention (e.g., more counseling sessions; interventions designed to address co-occurring psychiatric disorders in addition to substance use disorder) or changes to drug court program structure (e.g., changes in application or delivery of sanctions) directly impacts graduation rates. Given regional differences in drug court programming, future research should include multiple drug courts from different regions to gauge generalizability of findings. 
  • For policy makers: More funding for drug courts and research into drug court programming can help clarify aspects of programming that are effective as well as aspects of programming that can be changed/altered to improve outcomes. Given that drug courts can improve rates of substance use and recidivism, improving graduation rates and successful engagement in these programs is likely to have broader benefits for the criminal justice system and society at large. 

CITATIONS

Randall-Kosich, O., Whitaker, D. J., Guastaferro, W. P., & Rivers, D. (2022). Predicting drug court graduation: Examining the role of individual and programmatic characteristics. Journal of Substance Abuse Treatment, 135, 108654. doi: 10.1016/j.jsat.2021.108654


Share this article

l

WHAT PROBLEM DOES THIS STUDY ADDRESS?

Substance use disorders are prevalent among incarcerated people in the United States. Drug courts offer an alternative to incarceration and an opportunity to receive substance use disorder treatment for certain individuals within the criminal justice system. Drug court participation is associated with reduced rates of recidivism (i.e., re-arrest) and drug use; however, a large subset (30-50%) of drug court participants do not complete their respective programs – also known as “graduation”.

Determining the factors that influence drug court completion is a complicated endeavor. Most of the existing research has focused on which characteristics of drug-court participants (i.e., individual-level variables) predict graduation from these programs. Generally, participants who are older, female, White (vs. non-White), employed, and more educated are more likely to graduate from drug court programs. On the other hand, individuals with more serious criminal histories and more severe substance use and mental health difficulties are less likely to graduate. That said, individuals with more severe needs or complex histories are less likely to be admitted to drug courts in the first place, potentially skewing these results. Additionally, there has been less research in samples with greater racial/ethnic diversity.

Research on how programmatic factors—characteristics of the programs, including requirements, treatment received, etc.— can also help inform strategies to improve graduation rates. A study combining results from 76 studies of drug courts found that, in general, more requirements were related to an increase in program effectiveness, whereas certain requirements (sanctions, fines) were related to a decrease in effectiveness. However, there has been less focus on treatment received during these programs. Focusing on program-level factors is important because these factors can be changed and tailored to individuals’ needs, whereas individual-level characteristics are not necessarily as amenable to change.

In this study, researchers examined both individual and program-level factors that may influence rates of graduation to inform drug court policies and practices.


HOW WAS THIS STUDY CONDUCTED?

This study involved naturalistic follow-up and tracking of graduation outcomes among 247 participations from a large felony-level drug court in a southeastern U.S. city. The drug court had two tracks: 1) “Track 1” was for higher-risk participants (i.e., severe SUD) and involved five treatment phases over 24 months; 2) “Track 2” was for moderate-to-lower risk participants (i.e., those with stable employment, stable housing, fewer felony convictions) and included four treatment phases over 21 months. Participants in both tracks had to attend all court programming and received sanctions for infractions including positive drug screens, violent/destructive behavior, or re-arrest. Sanctions included treatment response (i.e., completing additional homework and/or increased intensity of treatment sessions), community service, jail, behavioral contracts, 30-day step-up (i.e., participants have 30 days to meet agreed-upon terms or are dismissed), and monetary fines and were applied using a matrix with consideration of contextual factors.

The data used in this study came from drug court records as well as arrest data from state records. Drug court records included data on days in program, type and attendance at treatment sessions, and number/type of sanctions issued. All participants completed two assessments upon admission to drug court, including an assessment of various indices of level of risk/need (e.g., criminal history, education/employment history, emotional/personal risk, and other factors) and an assessment of substance use history. Criteria for graduation from drug court included completing all program phases, having no positive drug screens for 6 months prior to graduation, completing an aftercare plan, and completing the final assessment (described earlier) at drug court exit.

The study’s primary purpose was to examine the extent to which individual-level and program-level factors (and their combination) predicted graduation from drug court. To accomplish this, researchers first examined which factors were individually related to graduation. Then they tested all of these factors at once, so that each variable’s unique effect could be examined while accounting for other variables initially related to graduation. This approach helps determine the independent effect of each variable, increasing confidence that the factor is causally related to drug court graduation.

The individual-level factors included variables related to individual substance use history and mental health, and the program-level factors included variables capturing amount and type of treatment received in addition to program requirements. The drug court studied was a large, felony-level drug court in that it included a diverse sample of moderate- and high-risk individuals, allowing the authors to examine a range of risk and participant characteristics.

The final sample included data from 247 drug court participants. Of the full sample, 77.7% were male, 61.8% were Black, 53.8% graduated high school, and 38.4% were employed; the average age was 38.9 years. 55.4% of participants reported cocaine as their drug of choice, 81.9% endorsed using a “criminogenic” drug of choice (i.e., crack cocaine/cocaine, heroin/other opiates, and methamphetamines/other amphetamines), and 50.9% reported using multiple substances. The average number of prior arrests was 17.2 total arrests and 9.9 felony arrests.


WHAT DID THIS STUDY FIND?

Less than half of participants graduated from drug court.

A total of 46% of participants graduated from drug court. Track 1 (i.e., higher risk track) participants were less likely to graduate from drug court than participants in Track 2 (38% vs. 55%).

More program factors were related to graduation than individual factors.

Only one individual-level factor was associated with graduation rates in the final combined analyses. Higher levels of emotional/personal risk (i.e., an index containing items assessing mental health, emotional management, and antisocial personality features) was associated with lower odds of graduation. Having a high score on this measure was associated with having a 44% decrease in the odds of graduation from drug court.

Multiple program-level factors predicted odds of graduation. Specifically, having a greater number of days in program and individual counseling sessions predicted higher odds of graduation, whereas having a greater number of sanctions decreased odds of graduation. Controlling for the number of days in the program, having more individual sessions was associated with a 20% increase in the odds of graduating. A greater number of jail and monetary sanctions predicted lower odds of graduation, decreasing the odds of graduation by 65% and 71%, respectively.


WHAT ARE THE IMPLICATIONS OF THE STUDY FINDINGS?

This study examined the effects of individual and programmatic factors on graduation from a large, urban drug court. Less than half of the drug court participants in the current study graduated from drug court, which is consistent with prior estimates of drug court graduation rates. In analyses examining both individual- and program-level predictors of drug court graduation, only one individual-level variable predicted graduation rates, whereas several program-level variables related to graduation.

Specifically, the only individual-level predictor of graduation, having higher levels of emotional/personal risk at baseline—which represents severity of current/past mental health conditions including current antisocial personality disorder—was related to lower likelihood of drug court graduation. Although personal mental health history is an individual variable, it does highlight an area that could be addressed when designing drug court programming. For instance, a pilot study of a drug court program that included a set of interventions addressing co-occurring mental health conditions (MISSION-CJ) showed promising outcomes. Participants who received this intervention had high levels of continued engagement in treatment, as well as reduced rates of recidivism (measured as nights in jail) and decreased co-occurring psychiatric and trauma symptoms. More rigorously designed studies comparing this type of intervention to traditional programming will be helpful in determining whether an increased focus on co-occurring psychiatric disorders can improve outcomes and graduation rates among drug court participants.

The focus on how programmatic features can influence graduation outcomes is important, as these features can be changed, whereas aspects of individuals’ backgrounds cannot. Program features that were associated with graduation in this study include those related to increased treatment – i.e., greater number of days in program and greater number of individual treatment sessions. By contrast, a greater number of punitive actions, particularly more sanctions (particularly jail and monetary), were associated with a decreased likelihood of graduation. However, this study used a design that was correlational, so it is unclear if these program features actually caused poorer outcomes or reflected individual differences. For instance, those who attended more treatment sessions may have just been more motivated and adherent to the programming itself, whereas those at higher individual risk (i.e., due to more severe SUD, co-occurring mental health difficulties, etc.) also may have been more likely to receive sanctions due to their behavior/non-adherence. While researchers did their best to address some of these alternative explanations for the findings with the study design, there were many factors for which they could not account given the data at their disposal. More research, ideally comparing drug courts with different intervention/sanction structures, is needed to elucidate the degree to which these program features impact graduation in a causal way.

The detrimental effect of sanctions has been found in several other studies. As noted, receiving sanctions can be indicative of other risk factors and can signify that a participant is nearing termination from drug court. Some researchers have started to study ways that sanctions can be implemented so that they are more likely to promote good outcomes. Specifically, sanctions that are “swift, certain, and fair” are associated with lower rates of substance use; however, their relationship with crime recidivism is less conclusive. Additional research examining ways that sanctions can be employed to promote positive behavior change is needed.


  1. This study’s design was observational, thus relationships between individual- and program-level variables and graduation rates are correlational in nature. Further research is needed to clarify the degree to which these variables directly impact graduation rates and whether there are relevant interactions between variables and/or additional intervening variables that could better explain these associations.
  2. The data included in this study was limited to what was available in records kept by the drug court, which in turn limited the amount of detail regarding type/quality of treatment, prior substance use history, prior treatment history, or time in recovery housing.
  3. The data in this study did not include information regarding the reasons why participants dropped out of drug court, which could better explain factors contributing to drop-out (e.g., mental health, motivational, other psychosocial factors, etc.).
  4. The sample only included individuals from one drug court. Although this drug court included a diverse set of participants representing a range of risk, findings cannot necessarily be generalized to other parts of the country.

BOTTOM LINE

This study examined the effect of individual and programmatic factors on a large, urban drug court. Less than half of participants in this drug court completed the intended duration of the treatment program and met its goals. When multiple individual-level and program-level predictors were examined at once, researchers found that having higher levels of co-occurring mental health/personality difficulties was associated with lower likelihood of graduating from drug court. At the program-level, a higher number of days in the program and more individual treatment sessions were associated with a greater likelihood of graduation, whereas a higher number of sanctions (particularly jail or monetary sanctions) were associated with lower graduation rates. Additional research is needed to clarify whether adding more individual counseling sessions, changing aspects of drug court programming to address mental health needs, and/or designing sanctions to be more effectively delivered, can improve outcomes.


  • For individuals and families seeking recovery: For individuals involved in the criminal justice system, drug court can be a viable option to avoid incarceration while also participating in programming to address a substance use disorder and reduce substance use. Having co-occurring mental health difficulties can reduce the likelihood of graduating from drug court; thus, supplemental mental health treatment or seeking additional individual intervention within a drug court program may better promote success. 
  • For treatment professionals and treatment systems: Participants in drug court are likely to benefit from continued support and interventions for co-occurring mental health conditions from treatment professionals. In addition to addressing baseline mental health needs, providers working with drug court participants may also assist patients with modifying behaviors that may interfere with successful engagement with drug court programming and ultimately graduation (e.g., motivation, other barriers to participation in treatment, etc.).
  • For scientists: These research findings suggest several possible avenues for further drug court research. Most existing research has been correlational or limited to pilot studies. More rigorously designed studies—randomized controlled trials or quasi-experimental studies with appropriate comparison groups and rigorous designs (e.g., propensity score matched comparison group)—are needed to determine if additional intervention (e.g., more counseling sessions; interventions designed to address co-occurring psychiatric disorders in addition to substance use disorder) or changes to drug court program structure (e.g., changes in application or delivery of sanctions) directly impacts graduation rates. Given regional differences in drug court programming, future research should include multiple drug courts from different regions to gauge generalizability of findings. 
  • For policy makers: More funding for drug courts and research into drug court programming can help clarify aspects of programming that are effective as well as aspects of programming that can be changed/altered to improve outcomes. Given that drug courts can improve rates of substance use and recidivism, improving graduation rates and successful engagement in these programs is likely to have broader benefits for the criminal justice system and society at large. 

CITATIONS

Randall-Kosich, O., Whitaker, D. J., Guastaferro, W. P., & Rivers, D. (2022). Predicting drug court graduation: Examining the role of individual and programmatic characteristics. Journal of Substance Abuse Treatment, 135, 108654. doi: 10.1016/j.jsat.2021.108654


Share this article