Printer Friendly

Optimizing the Effectiveness of Correctional Programming: The Importance of Dosage, Timing, and Sequencing.

May 2019

Key Points

* Programming dosage should be calibrated to risk, with higher-risk prisoners receiving longer, more intensive interventions.

* As program participation increases, recidivism generally decreases. Recidivism outcomes are significantly better when prisoners participate in multiple interventions or spend much of their imprisonment in programming.

* Back-loading programming closer to release from prison has been associated with better recidivism outcomes.

* Program seguencing may be effective for those who participate in multiple interventions.

Following the infamous conclusion drawn in the 1970s that "nothing works" for correctional populations, (1) a large number of studies have demonstrated there are some effective interventions. Dubbed the "what works" literature, these studies have helped identify programs that have proved to be successful in improving outcomes such as recidivism, prison misconduct, and post-release employment. Examples of effective correctional interventions include substance abuse treatment, cognitive-behavioral therapy (CBT), sex offender treatment, and education and employment programs. (2)

While the "what works" literature has helped reveal which interventions tend to achieve positive outcomes, research also suggests the efficacy of programming may vary across individuals. Whereas CBT programming may be effective for some participants, it may not produce desistance from crime for others. To be sure, knowing what works is crucial. However, to achieve better outcomes for a larger segment of the correctional population, it is also crucial that we determine what works best for whom and under what circumstances.

To that end, this report considers a number of questions relating to the dosage, timing, and sequencing of correctional programming. For example, how much programming is needed to achieve desistance? Does it matter when individuals participate in programming during their time under confinement or community supervision? Is it more effective to front-load programming toward the beginning of one's sentence, or is it better to back-load it closer to the time of release? If an individual can participate in multiple interventions, does the sequence in which he or she participates in these interventions matter? Is it better to participate in education programming before entering a CBT intervention, or vice versa?

Before considering these questions, however, it is important to briefly review the risk-needs-responsivity (RNR) model, which has become the prevailing paradigm that guides the delivery of programming for correctional agencies across North America. The RNR model identifies who should be treated (risk), what areas should be treated (needs), and how treatment should be delivered (responsivity). More specifically, because correctional resources are often scarce, the risk principle suggests we can enhance the benefits from treatment by focusing on higher-risk offenders. While the needs principle holds that interventions that target criminogenic needs (dynamic risk factors) are more likely to decrease recidivism because changes can be made in these factors, the responsivity principle proposes that programming should be tailored to offenders' learning styles, abilities, and strengths. (3)


Existing research has shown that programming dosage (i.e., how much programming a person receives) should be calibrated to risk. (4) This evidence implies that delivering the right amount of programming hinges on accurately assessing risk. In general, as risk increases, so does the amount of programming needed to bring about desistance. Therefore, while some lower-dosage interventions may work for lower-risk individuals, they are less likely to be effective for those who are higher risk. Instead, higher-risk offenders often require longer, more intensive interventions. (5) As we see below, however, the relationship between dosage and desistance depends, to some extent, on how dosage has been measured and the type of intervention.

Most of the dosage research has looked at the number of hours participants spent in programming, and these studies have focused on two types of interventions: CBT and substance abuse treatment. The findings from the evaluations of CBT programs indicate that greater dosages are associated with better recidivism outcomes. (6)

In their study of 620 Canadian offenders, Guy Bourgon and Barbara Armstrong examined recidivism outcomes among four different groups that varied according to the extent to which they participated in the programs: (1) untreated, (2) five weeks/100 hours of treatment, (3) 10 weeks/200 hours, and (4) 15 weeks/300 hours. Bourgon and Armstrong found that as the dosage of CBT increased, recidivism decreased. (7) Similarly, in their study on 13,676 offenders who participated in a variety of treatment programs, Chris Lowenkamp, Ed Latessa, and Alex Holsinger reported better recidivism outcomes for higher-risk offenders when they received more treatment. (8) More recently, a study examined the relationship between hours of participation in CBT programming and recidivism risk among 689 male offenders from Ohio. For the higher-risk offenders, higher dosages of treatment yielded better recidivism outcomes. (9)

While some lower-dosage interventions may work for lower-risk individuals, they are less likely to be effective for those who are higher risk.

Prior research on substance abuse treatment has shown that more programming does not always lead to better outcomes because there is a point at which longer durations of treatment can produce diminishing returns. In their evaluation, Harry Wexler, Gregory Falkin, and Douglas Lipton reported that as time in a substance abuse treatment program increased, so, too, did the time until rearrest. But treatment participants recidivated more quickly after having been in the program longer than 12 months, which is when, according to the researchers, participants may have become disillusioned and reduced their involvement in the program. (10)

Consistent with this research, an evaluation of chemical dependency treatment for Minnesota prisoners found that increased treatment time appeared to lower the risk of recidivism, but only up to a point. Although short-term (90 days) and mediumterm (180 days) programs had a statistically significant impact on all three recidivism measures, no significant effects were found for long-term (365 days) programming. (11) Similarly, an evaluation of substance abuse treatment for Florida prisoners also found that more time spent in substance abuse treatment did not significantly reduce recidivism. (12)

Much of the existing literature on correctional treatment dosage has measured it as the number of hours, days, or weeks that individuals participate in an intervention. A study on Minnesota prisoners explored the dosage-recidivism relationship with another measure--the number of correctional interventions in which prisoners had participated. Examining more than 55,000 offenders released from Minnesota prisons, the study reported that nearly one-third had been "warehoused," meaning they did not participate in any programming while in prison. The results showed that warehousing prisoners significantly increased recidivism by 13 percent. On the other hand, participation in effective interventions significantly reduced recidivism. The size of the reduction was greater for individuals who were involved in multiple effective interventions and, thus, presumably had higher dosages of programming. (13)

Another study on Minnesota prisoners looked at dosage in terms of the number of days in programming relative to the total imprisonment period. The 1,879 prisoners examined were, on average, in programming for 36 percent of their time in prison. Put another way, for every 100 days in prison, these inmates were in programming for 36 days. The findings showed a reduction in recidivism when prisoners spent more of their time in programming. In particular, a 1 percentage point increase in programming resulted in a 22 percent decrease in recidivism. (14)


The evidence from the dosage literature suggests recidivism outcomes would be better if inmates were involved in programming during much of their imprisonment. As noted above, however, many prisoners are warehoused, and warehousing exacerbates both recidivism and employment outcomes. (15) Moreover, when prisoners can participate in programming, the dosage is often insufficient due to a lack of resources. If resources are limited, are there still ways to optimize programming's beneficial effects? That is, are there conditions under which interventions are more likely to be effective?

The timing of correctional interventions--or, more specifically, the point at which prisoners enter or exit interventions during their imprisonment--may have important implications for the effectiveness of programming. Indeed, the timing concept implies that the success of an intervention hinges, at least in part, on when it is delivered to offenders. The prisoner reentry literature has maintained that the reentry process should begin as soon as individuals enter prison, (16) which would entail early involvement in programming. Still, some research suggests that programming may be more beneficial if individuals participate in an intervention toward the end of their confinement period, as opposed to the beginning. (17)

If resources are limited, are there still ways to optimize programming's beneficial effects?

Among the few prior studies that have examined whether timing matters for programming, the results suggest that interventions taking place closer to an individual's release from prison tend to have a greater impact in reducing recidivism. In addition to possibly reflecting the benefits of a continuum of care from prison to the community, shorter intervals between program exits and release dates may help better preserve the positive effects of programming.

Although prison visitation is seldom considered a type of correctional programming, it has been associated with reduced recidivism. By providing prisoners with pro-social support, prison visitation presumably addresses antisocial peers--a major recidivism risk factor. Prior studies from Florida (18) and Minnesota (19) have found that visits closer to prisoners' release dates exerted a greater influence in reducing reoffending.

More recently, an evaluation of a substance abuse treatment program for Florida prisoners examined the timing of program completion to release from prison. Overall, this study showed that substance abuse treatment significantly lowered recidivism. Focusing on the length of time between treatment completion and release from prison, the researchers found significantly better recidivism outcomes for one of the treatment modalities they examined when participants completed the program closer to their date of release. (20)

The aforementioned study on 1,879 Minnesota prisoners also examined the effects of program timing on recidivism. The results showed the point at which prisoners entered programming did not have a direct, significant effect on reoffending. Instead, the point at which prisoners exited programming, particularly in relation to the overall length of their imprisonment, had a greater impact on recidivism. When prisoners exited programming closer to their release from prison, recidivism outcomes were generally better.

Exiting programming closer to the time of release may help prisoners better retain the positive effects that interventions have on their post-release behavior. For this same reason, the absence of a significant association between program entry and recidivism may reflect the possibility that an intervention's impact can fade over time. Accordingly, while earlier involvement in programming may not significantly reduce recidivism, it leads to greater participation in interventions, and it could have a significant impact on a more proximate outcome: prison misconduct. (21)


Although the evidence suggests that back-loading programming will likely lead to more favorable outcomes, the truth is the vast majority of prisoners have multiple risk factors that make desistance difficult to achieve. To successfully address these risk factors, many prisoners need to participate in multiple interventions to significantly reduce their recidivism risk.

But does the order, or sequence, in which individuals participate in these interventions matter? Moreover, are there certain combinations of programming that optimize outcomes? For example, is CBT more effective when it is paired with education programming or substance abuse treatment?

Existing research has yet to empirically examine whether program sequencing matters for outcomes such as prison misconduct, recidivism, or post-release employment. Some scholars have theorized, however, that the order in which offenders participate in interventions may affect programming's overall effectiveness. For example, Donna Mailloux and colleagues suggest:
It may be useful for offenders to complete a program such as cognitive
skills (which introduces basic elements associated with
cognitive-behavioral therapy as well as concrete suggestions as to how
to apply these principles to everyday situations) prior to completing
more intensive therapeutic programs. (22)

Alternatively, another approach would be to first address the most influential risk factors, such as criminal thinking or antisocial peers. After completing, say, a CBT program, it may then be possible to address more moderate risk factors such as education, employment, or substance abuse.


More research is needed, of course, to determine whether these and other sequencing strategies might produce better outcomes. It is important to emphasize, however, that optimizing the delivery of correctional programming hinges on accurate assessments of risk, needs, and responsivity. Risk assessment not only tells us who to prioritize for programming but also what their dosage should be. Needs assessment indicates which areas we should target for programming, while responsivity assessment should be able to identify the interventions that would likely be most beneficial for an individual. (23)

The literature shows that greater participation in interventions would likely yield better recidivism outcomes, and a good way to achieve that is to get prisoners involved in programming shortly after their admission to prison. Earlier involvement may not lead directly to less recidivism, but it can increase the extent to which prisoners participate in programming, which lowers reoffending.

But even if prisoners want to participate in interventions, access to programming can be limited due to a scarcity of resources. Indeed, many prisoners do not participate in any programming. Among those who do, the overall dosage rate found in the study on Minnesota prisoners was 36 percent, (24) which suggests prisoners are not involved in programming for a substantial portion of their time in prison.

Increasing the quantity of programming would raise the overall dosage percentage, which would, in turn, likely yield better recidivism outcomes. Without an increase, however, corrections agencies should attempt to back-load programming if recidivism reduction is the goal. For example, if prisoners can participate in only one intervention due to shorter imprisonment periods or resource limitations, it should be reserved toward the end of confinement to maximize the effect on recidivism. Even among those who can participate in multiple interventions, participation in one of the interventions should be set aside toward the end of imprisonment to achieve the best public safety outcomes.

About the Author

Grant Duwe is an academic adviser to AEI for criminal justice reform. He is also the research director for the Minnesota Department of Corrections, where he develops and validates risk assessment instruments, forecasts the state's prison population, and conducts research studies and program evaluations. Duwe has published more than 60 articles in peer-reviewed journals on a wide variety of correctional topics, and he is a coauthor of the book The Angola Prison Seminary: Effects of Faith-Based Ministry on Identity Transformation, Desistance, and Rehabilitation.


(1.) R. Martinson, "What Works? Questions and Answers About Prison Reform," Public Interest 34 (1974): 22-54.

(2.) Grant Duwe, The Use and Impact of Correctional Programming for Inmates on Pre- and Post-Release Outcomes, US Department of Justice, Office of Justice Programs, National Institute of Justice, June 2017,

(3.) D. A. Andrews, James Bonta, and J. Stephen Wornith, "The Recent Past and Near Future of Risk and/or Need Assessment," Crime & Delinquency 52, no. 1 (January 2006): 7-27,

(4.) Chris T. Lowcnkamp, Ed J. Latcssa, and Alex Holsingcr, "The Risk Principle in Action: What Have Wc Learned from 13,676 Cases and 97 Correctional Programs?," Crime & Delinquency 52 (2006): 77-93.

(5.) K. G. Spcrbcr, E.J. Latcssa, and M. D. Makarios, "Examining the Interaction Between Level of Risk and Dosage of Treatment," Criminal Justice and Behavior 40 (2013): 338-48.

(6.) Mark W. Lipsey, Nana A. Landenberger, and Sandra J. Wilson, "Effects of Cognitive-Behavioral Programs for Criminal Offenders," Campbell Systematic Reviews 6 (August 2007),

(7.) Guy Bourgon and Barbara Armstrong, "Transferring the Principles of Effective Treatment into a 'Real World' Prison Setting," Criminal Justice and Behavioral (February 2005): 3-25,

(8.) Lowenkamp, Latessa, and Holsinger, "The Risk Principle in Action."

(9.) Sperber, Latessa, and Makarios, "Examining the Interaction Between Level of Risk and Dosage of Treatment."

(10.) Harry K. Wexler, Gregory P. FaUdn, and Douglas S. Lipton, "Outcome Evaluation of a Prison Therapeutic Community for Substance Abuse Treatment," Criminal Justice and Behavior 17, no. 1 (March 1990): 71-92,

(11.) Grant Duwe, "Prison-Based Chemical Dependency Treatment in Minnesota: An Outcome Evaluation," Journal of Experimental Criminology 6, no. 1 (March 2010): 57-81,

(12.) Samuel Scaggs ct al., An Assessment of Substance Abuse Treatment Programs in Florida's Prisons Using a Random Assignment Experimental Design, National Institute of Justice, April 2016,

(13.) Grant Duwe and Valeric Clark, "The Rehabilitative Ideal Versus the Criminogenic Reality: The Consequences of Warehousing Prisoners," Corrections: Policy, Practice and Research 2, no. 1 (October 2016): 41-69,

(14.) Grant Duwe, "The Effects of the Timing and Dosage of Correctional Programming on Recidivism," Journal of Offender Rehabilitation 57, no. 3-4 (2017): 256-71,

(15.) Grant Duwe and Valerie Clark, "Nothing Will Work Unless You Did: The Predictors of Post-Prison Employment," Criminal Justice and Behavior 44 (2017): 657-77.

(16.) Nancy La Vigne et al., Release Planning for Successful Reentry: A Guide for Corrections, Service Providers, and Communitv Groups, Urban Institute, September 2008,

(17.) Duwe, "The Effects of the Timing and Dosage of Correctional Programming on Recidivism."

(18.) W. D. Bales and D. P. Mcars, "Inmate Social Tics and the Transition to Society: Docs Visitation Reduce Recidivism?," Journal of Research in Crime and Delinquency 45, no. 3 (2008): 287-321,

(19.) Grant Duvvc and Valeric Clark, "Blessed Be the Social Tic That Binds: The Effects of Prison Visitation on Offender Recidivism," Criminal Justice Policy Review 24, no. 3 (2013): 271-96,

(20.) Scaggs ct al., An Assessment of Substance Abuse Treatment Programs in Florida's Prisons Using a Random Assignment Experimental Design.

(21.) Duwe, "The Effects of the Timing and Dosage of Correctional Programming on Recidivism."

(22.) Donna L. Mailloux et al., "Dosage of Treatment to Sexual Offenders: Are We Overprescribing?," International Journal of Offender Therapy and Comparative Criminology 47, no. 2 (April 2003): 171-84,

(23.) Grant Duwe and K. Kim, "The Neglected 'R' in the Risk-Needs-Responsivity Model: A New Approach for Assessing Responsivity to Correctional Interventions," Justice Evaluation Journal 1, no. 2 (2018): 130-50,

(24.) Duwe, "The Effects of the Timing and Dosage of Correctional Programming on Recidivism."

By Grant Duwe
COPYRIGHT 2019 The American Enterprise Institute
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2019 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Duwe, Grant
Publication:AEI Paper & Studies
Geographic Code:1U4MN
Date:May 1, 2019
Next Article:America's Invisible Felon Population: A Blind Spot in US National Statistics.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters