Cognitive–Behavioral Intervention for Children with Emotional and Behavioral Disturbances
*Note: The National Mentoring Resource Center makes these “Insights for Mentoring Practitioners” available for each program or practice reviewed by the National Mentoring Resource Center Research Board. Their purpose is to give mentoring professionals additional information and understanding that can help them apply reviews to their own programs. You can read the full review on the CrimeSolutions.gov website.
In considering the key takeaways from the research on this program that other mentoring programs can apply to their work, it’s useful to reflect on the features and practices that might have influenced its rating as “Promising” (that is, a program that shows some evidence that it achieves justice-related goals when implemented with fidelity).
1. This program represents a very thoughtful approach to integrating a unique, standalone intervention into a youth mentoring context.
Mentoring practitioners often borrow ideas or curricula from related fields and disciplines when developing mentoring interventions designed to address specific youth challenges or behaviors. Certainly the reviews of programs conducted to date by the National Mentoring Resource Center are filled with examples of such programs. There can be tremendous potential in applying successful approaches from one context to the work of a mentor. That certainly appears to be the case for this program, which brought many of the principles of the highly-successful cognitive behavioral approach to therapy to mentoring relationships for youth.
Cognitive behavioral therapy has been shown to produce meaningful results for adults and youth in relatively short periods of time and that short-term focus is seen in this program, which offered youth only 12 weeks of mentoring⎯a far cry from the school- or calendar-year focus of many mentoring programs. And luckily for practitioners, the developers of this intervention even merged this psychological intervention into a mentoring context in two different ways: A traditional one-to-one program and a group program (although, unfortunately, they ran these programs separately at different times and did not compare the results between the two approaches).
Programs looking to integrate principles or ideas from other interventions or disciplines can find inspiration in the table included on page 205 of the 2009 evaluation of the program in Child & Family Behavior Therapy (you can view a full text version of the article here). This table illustrates how the developers compared and contrasted the practices of cognitive behavioral therapy and typical youth mentoring (group, in this case) across several key dimensions (time, activities, length of meeting, parent involvement, relationship focus, level of training, transportation, and employment status of provider). The table also shows how they eventually merged these practices into the hybrid model that they implemented and studied. This illustrates how practitioners can go about thoughtful integration of one intervention with another. Rather than just taking a curriculum or a set of ideas from one context and assuming it can work in another, these developers looked at all of the factors that made one intervention work and thought carefully about how those practices could be brought over or modified for a mentoring context. Any program looking to bring new ideas to a mentoring program environment should plan on making a table like this to show how they think the interventions will mesh and what might need tweaking to make it work in practice.
The same journal article also offers a clear breakdown of the content of the 12 weeks of the program illustrating how key cognitive behavioral concepts, such as relaxation techniques, perspective taking lessons, social problem-solving, and tips for managing challenging or embarrassing situations, were woven into the mentor-mentee interactions. In all, this program, in both of its iterations, really shows how to blend interventions in a thoughtful way.
2. This program shows the value of testing multiple approaches to addressing the same issue for youth.
As noted above, this program was offered in two formats designed to achieve similar goals. In the earlier one-to-one iteration of the program (the results of which were published in 2006) the program focused on not just improving the behavior of mentees but also specifically tested whether the intervention helped parents and caregivers by increasing their feelings of support and reducing their stress around their child’s circumstances and behavior. The idea being that if the mentors could improve the child’s behavioral issues, that that might reduce stress in the home and lead to even greater outcomes. And in fact, that’s exactly what they found. Findings supported their idea that reducing parent stress in the home was a pathway by which the mentoring was able to benefit behavioral outcomes for youth. So this version of the program attempted to address these youths’ needs on two fronts: direct work with the mentor and indirect impact on the home environment by reducing parent stress, increasing their feelings of support, and bringing more harmony to the home environment. More mentoring programs should explore how the direct work with a mentor can translate to or be reinforced in the home environment and whether the impact of the mentoring extends to others in the family, even if indirectly.
The later published evaluation of the program in 2009 put a different spin on this work: a group format, which offered additional adults to teach and model the behaviors taught in the curriculum and more opportunities for practicing newly learned strategies and skills with peers in a variety of situations. The program was designed with the aim of ensuring that mentees still got the individual attention they needed, but with more chances to bring what they were learning to life in real world contexts (most notably actual interactions with peers).
As with the one-to-one version of the program, the evaluation of the group model also found evidence of favorable impacts on the youth’s behavior, including both internalizing and externalizing problems. But unlike the one-to-one model, this version did not show evidence of improving parent feelings of stress. This difference in findings across the two studies emerged in spite of the fact that in both versions of the program, mentors spent some time after each session speaking with parents about their child’s progress and discussing the activities and lessons. It is unclear as to why the group program did not improve parent stress (perhaps the group format resulted in less specific information-sharing about each parent’s particular child). Unfortunately, the 2009 study did not look at one key factor that the 2006 study did: Parent reported perceptions of social support in managing their child’s behavior. In the one-to-one version of the program, parents in the program group reported feeling more supported and less stress (presumably as a result of that social support). But the later published study simply didn’t ask if they felt more supported, so we don’t know if parents felt that way or not.
Perhaps the ideal version of this program is a “best of both worlds” one that combines the group structure to provide more mentoring, more peer support, and more ability to use what’s being taught, combined with additional emphasis on that parent support aspect and reducing the stress in the home. One can imagine that such a combination would perhaps maximize the impact of the program. That’s yet another iteration of the program that would have to be studied in the ways that these were. But there seems to be enough evidence here that mentoring programs explicitly serving youth with emotional and behavioral challenges would be well advised to consider how they can support parents and reduce that in-home stress.
More generally, mentoring programs that are just getting started or want to incorporate a new concept into a mentoring context may want to try different approaches simultaneously and study them to see which works best and the different reasons or unique pathways that explain their success or failure. That approach can help practitioners refine the idea of the program and eventually optimize services and the potential impact the mentoring can have.
3. The use of a “token economy” is an intriguing idea in a mentoring context.
In the evaluation of the group mentoring version of the program it is briefly mentioned that mentors instituted a simple “token economy” to reward mentees for their positive behavior and for instances of applying the recently-taught lessons of the program correctly in real-life situations. These rewards, or “points,” could then possibly be exchanged for other rewards, such as playing a particular game or going on a fun outing. The idea of reinforcing positive behavior in this way is one that mentoring programs that are looking to influence behavior in the real world might consider. There is growing evidence that this kind of “gamification” can improve behavior in the context of other types of interventions, perhaps most notably in Positive Behavioral Interventions and Supports. There are concerns that this gamified approach reduces intrinsic motivation and leaves youth simply seeking the external reward. But for a mentoring intervention designed to get youth adopting and practicing targeted behaviors with some regularity, reinforcing that behavior through some kind of token economy might be a nice additional incentive that can help make using the desired strategies a routine habit over time.
4. When evaluating a program for youth with substantial needs, it’s always a good idea to control for other services.
One of the key things that the evaluators did in both of the studies mentioned here is control statistically for the volume of other services that mentored youth and their families were receiving. These researchers understood that these youth would be getting lots of other support and other interventions on top of what their program was offering (e.g., medication, case management, out-patient therapy, etc.). So in an effort to better isolate the effects of the program, they also took into account the volume of these other services and supports. It seems as though those other services did not influence the findings in either study, but the researchers are quick to note that their method of controlling still leaves room for that possibility.
To try and quantify the volume of other support mentees were getting, the researchers simply totaled up the number of hours of support the youth was receiving and noted whether the youth was getting medication. This composite approach made it easier to control for the potential influence of these other services, but it lacked nuance and specificity. They did not track (or if they did, did not utilize information on) whether the mentee started, stopped, or switched medication during the program, nor did they control separately for the type of treatment approaches they were engaged in. As a result, there is some uncertainty around how the mentoring program may have different impacts for youth getting various other supports or changing their medications. This illustrates just how hard it can be to control for the myriad other factors that may influence a mentoring program’s success. Practitioners and researchers are often left executing a simple-yet-incomplete model of what those other factors might be, weighing how much information they can reasonably collect with the knowledge that they must control for external factors somehow. This program, and the researchers approach to evaluating it, offer a good example of how to split that difference in a way that is doable while also being upfront about the approach not being perfectly complete.
For more information on research-informed program practices and tools for implementation, be sure to consult the Elements of Effective Practice for Mentoring™ and the "Resources for Mentoring Programs" section of the National Mentoring Resource Center site.