ADHD TREATMENT IN COMMUNITY SETTINGS - IMPACT ON EDUCATIONAL OUTCOMES

 
In prior issues of Attention Research Update, detailed results from the Multi-site, Multi-modal Treatment Study for Children with ADHD (the MTA study) have been presented.  The results of this groundbreaking study -- the largest and most comprehensive treatment study for ADHD ever conducted -- clearly established that carefully conducted medication therapy and intensive behavioral treatments are effective interventions for children with ADHD.  The combination of medication and behavior therapy was slightly more effective overall than either treatment used in isolation, and over two-thirds of children who received combined treatment had symptom scores for ADHD and oppositional behavior that fell within the normal range after 14 months.
 
Although these results are encouraging, it is important to recognize that they demonstrate the success rate that is possible to achieve when medication treatment and behavior therapy are provided in a rigorous and comprehensive manner.   Unfortunately, the level of care that children with ADHD receive in typical community settings may often fall short of this standard. As a result, even though a child may receive treatments with well-established efficacy, the results may fall short of what is obtained in carefully conducted research studies.  For example, in the MTA study, some children were treated as usual in their communities so that the effect of the routine community treatment could be compared to the rigorously conducted treatments provided in the study.  Results indicated that community-treated children did not do nearly as well -- only 25% had ADHD symptoms and oppositional behavior that fell within the normal range after 14 months.  This is far lower than the 68% success rate for children who received combined treatment through the study.
 
Despite the wealth of information contained in MTA study results published so far, there remains one limitation: information on "real-world" academic outcomes -- a domain in which children with ADHD often struggle considerably -- has not been provided.  For example, in addition to learning about treatment impact on core ADHD symptoms, or even on academic achievement as measured by standardized testing, it would be important to know how treatment impacts such important academic outcomes as receiving special education services, being expelled or suspended from school, and having to repeat a grade.  These outcomes are critically important in a child's life and of great concern to parents.
 
What do we know about how treatment for ADHD in the community impacts important outcomes such as these?
 
A recently published study conducted in 3 elementary schools in southeastern Virginia provides some very important -- and alarming -- data on this critical "real-world" question (LeFever, G.B. et al. (2001). Parental perceptions of adverse educational outcomes among children diagnosed and treated for ADHD: A call for improved school/provider collaboration. Psychology in the Schools, 39, 63-71).
 
In this study, parents of all 1644 children in 3 schools received a survey asking whether their child had ever been diagnosed with ADHD, and, if so, what types of treatments their child was receiving.  Parents also were asked how they perceived their child's performance at school, whether their child was receiving special educational services, whether their child had ever been suspended or expelled, and whether their child had ever been retained.  The survey was completely anonymous and responses were obtained from over 60% of the sample.  This response rate is not quite as high as one would like, but is typical of what is generally obtained in this type of survey research.

 
RESULTS
 
How Frequently were Children Diagnosed with ADHD?
 
As you may be aware, the commonly reported prevalence rate for ADHD is between 3 and 5% of the population of school-age children.  In this sample, however, 17% of the students had received an ADHD diagnosis according to their parents.  Rate of reported diagnosis for boys was significantly higher than for girls (28% vs. 11%).  This is an enormously high rate of diagnosis -- several times higher than the commonly reported prevalence rates.
 
There are several possible factors that may have contributed to this high rate of diagnosis.
First, it is possible that some parents reported that their child had been diagnosed with ADHD when this had not occurred, although this seems unlikely. Another possibility is that parents with a diagnosed child were more likely to return the survey because it asked specifically about a condition that their family was dealing with.  This would have would have made the rate of diagnosis in this sample higher than it would have been if all parents returned the surveys.  It is also possible that ADHD may be over-diagnosed in this community, as experts believe that both over- and under- diagnosis of ADHD occurs, and that this can vary from one community to the next.
 
Finally, these data raise questions about the accuracy of the 3-5% prevalence rate for ADHD that is generally reported and suggests that actual prevalence rates may be higher.  (In this regard, it is worth noting that higher rates than the generally accepted 3-5% figure have also been reported in other studies.)
 
What Types of Treatment Were Diagnosed Children Receiving?
 
Children diagnosed with ADHD were receiving a range of services to treat their condition.  Most of the children -- 84% -- received medication to treat the disorder, although only 56% were receiving medication during the school day.  More than half of the children -- 57% -- were receiving behavioral interventions. And, nearly half of the children -- 47% -- were receiving both medication and behavioral treatment.  Only 27% received medication treatment alone; only 10% received behavioral treatment alone; and only 16% were receiving no treatment at all.
 
These results seem encouraging in that the vast majority of children diagnosed with ADHD were receiving treatments for which extensive empirical support exists -- i.e. medication and behavior therapy -- and many diagnosed children were being treated using both approaches.  This is certainly consistent with treatment guidelines recently published by the American Academy of Pediatrics. 
 
Several important differences in treatment services received by black and white students were evident.  Although medication use did not differ for black and white students diagnosed with ADHD, whites were significantly more likely to receive behavioral interventions (69% vs. 43%) and were also more likely to receive combined treatment (69% vs. 31%).  It is also interesting to note that among students without health insurance, none were receiving behavioral treatment.  Instead, they were treated exclusively with medication.  Data on treatment differences according to gender were not presented.
 
Adverse Educational Outcomes for Children with ADHD
 
Eighty-four percent of parents with ADHD children reported that ADHD affected their child's school performance and 40% believed the school was not providing adequate services to meet their child's needs.  The association between ADHD and adverse education outcomes were as follows:
 
* Children with ADHD were about 5 times as likely as other children to be receiving special education services;
 
* Children with ADHD were about 7 times as likely as other children to have been suspended or expelled;
 
* Children with ADHD were about 3 times as likely as other children to have repeated a grade.
 
(Note: Special education services should not necessarily be considered an adverse outcome as it is intended to provide additional education support and services for children with legitimate needs.  Many children with ADHD also have co-occurring learning disabilities and would require special education services for this reason alone.)
 
These results are striking. They indicate that, despite the fact that the vast majority of ADHD children in these schools were receiving empirically supported treatments, they were still faring far worse than their peers on these important indicators of academic functioning.  What is perhaps even more striking is that when the researchers compared children with ADHD who were receiving combined treatment -- the approach found to be most effective in the MTA study -- with children who were receiving no treatment at all, they found no differences in these outcomes.  In other words, they found no evidence that combined treatment reduced the likelihood that a child with ADHD would require special education services, would be expelled, or would have repeated a grade.
 
(Note: The authors do not specify whether parents were asked to indicate whether adverse outcomes occurred before or after their child began receiving treatment for ADHD, and it is possible that these figures look worse than they really are because at least some of the adverse outcomes occurred before treatment began.  It seems unlikely, however, that this could fully account for the substantial discrepancy in negative educational outcomes between children with ADHD and their peers.  The authors do recognize that this is a limitation of the study, however, and that information on the timing, intensity, and quality of ADHD services delivered might have shed additional light on the generally poor outcomes associated with ADHD that they report.)

 
SUMMARY AND IMPLICATIONS
 
These results are sobering.  They indicate that -- at least as practiced in this community -- children receiving treatment for ADHD continued to experience adverse educational outcomes at a rate that far exceeded that of other children.  In fact, there was no indication that treatment had any beneficial effect on the educational outcomes for children with ADHD that were examined.
 
How can this be?  It is tempting to attribute these findings to something unique to this particular community and hope that it is not representative of the effectiveness of ADHD treatment services provided in other communities.  Although such a possibility cannot be definitively ruled out, it does not seem to be a particularly plausible explanation.  In fact, it appeared that a larger percentage of students diagnosed with ADHD in this community were receiving empirically supported treatments than is often found.
 
The authors suggest, correctly, I believe, that their results raise important questions regarding "treatment effectiveness as opposed to clinical efficacy". Clinical efficacy examines the benefits of a treatment in a controlled setting, such as the treatment provided in the MTA study.  Treatment effectiveness, in contrast, is concerned with the actual benefits of a treatment as it is actually delivered in community settings.  As the results of this study make clear, when it comes to treatments for ADHD, there may be a substantial difference between clinical efficacy and treatment effectiveness.
 
It would be both incorrect and potentially harmful to interpret these results as indicating that medication treatment and behavioral interventions for ADHD are not effective.  In fact, we know from the MTA study -- and from other studies as well -- that when these treatments are carefully delivered, they can be very effective over a sustained period of time.  Instead, careful attention needs to be directed to learning why treatment outcomes that are typically attained in community settings are so much less positive than what is possible to achieve.
 
There are several possible reasons for this.  Regarding behavioral treatment, it is important to note that the behavioral intervention provided in the MTA study -- although effective -- is probably beyond what could ever be routinely available in most communities.  It included a combination of an intensive summer camp treatment experience, extensive parent training, and an extremely rigorous classroom-based behavior management system delivered by highly trained paraprofessionals.  Noted ADHD expert Dr. Russell Barkley has suggested that this form of intensive treatment is beyond what could reasonably be implemented on a large scale.  Whether less intensive behavioral treatment could yield equally positive results to those obtained in the MTA study thus remains somewhat unclear.
 
The issues surrounding medication treatment, in contrast, do not seem nearly so difficult to surmount.  In the MTA study, there were several aspects of medication treatment that, although not routinely done, may not be that difficult to consistently implement.  First, children received an initial trial that included a full range of doses. Systematic feedback was obtained from parents and teachers to determine the most effective dose.  Second, if a positive response was not obtained on any dose tested, another stimulant medication was tested in the same way.  It was only after a child failed to show a sufficiently positive response to any dose of either 2 or 3 different stimulants that a different class of medication, such as an antidepressant, was tried.  Finally, children's functioning was monitored on a monthly basis, an effort that included receiving information directly from teachers.  When it was evident that a child's symptoms were no longer being managed effectively, an adjustment to either dosage or medication type was implemented in an effort to bring the child's symptoms back under better control.  During the 14-month study, the majority of children receiving medication required one or more such adjustments.
 
Although a reasonable variant of MTA medication treatment procedures would not be difficult nor particularly time consuming to implement, there is little indication that they are routinely employed in community settings.  This is truly unfortunate, because the MTA results suggest that implementing these procedures could make an important difference in treatment outcomes.  For example, just making sure that physicians received regular feedback from teachers about a child's behavior and academic performance at school would be enormously helpful for deciding whether treatment was proving effective or whether adjustments to treatment were necessary.
 
The authors of the current study conclude by noting that one of the most important ways to increase the effectiveness of treatment for ADHD is to "...improve collaboration between physicians and school professionals" and that "taking proactive steps toward meaningful school/community collaboration is essential if we are to reduce the public health and educational crisis surrounding ADHD identification, treatment, and outcomes."  One hopes that this important call-to-action will be heeded.

Although it was previously believed that the vast majority of children with ADHD would simply "outgrow" the disorder during adolescence, this is now known to be incorrect.  Most children with ADHD continue to struggle with the condition during the adolescent years, and, even when they no longer meet full diagnostic criteria for the condition, often experience symptoms that contribute to difficulty in such diverse areas as school, peer relations, family relations, and self-esteem.
 
Unfortunately, relative to the amount of treatment research conducted with children, the available treatment studies of adolescents with ADHD are limited.  Stimulant medication treatment has been shown to be effective for adolescents. And, longer-acting medications such as Concerta and AdderallXR, which eliminate the need for in-school dosing, may help reduce compliance problems common to this age group.
 
As with children, however, medication treatment alone may not be adequate, particularly in those instances where co-occurring behavior difficulties such as those associated with Oppositional Defiant Disorder (ODD) or Conduct Disorder (CD) are also present. In such instances, the level of parent-teen conflict that is present may require direct work with families to reduce conflict and improve family functioning.   (For a discussion of ODD and CD click here.)

Recently, an extremely well-conducted study comparing two family-therapy approaches for adolescents with ADHD and ODD was published in the Journal of Consulting and Clinical Psychology (Barkley, RA., et al., (2001). The efficacy of problem-solving communication training alone, behavioral management training alone, and their combination for parent-adolescent conflict in teenagers with ADHD and ODD. JCCP, 69, 926-941).
 
Participants in this study included 97 adolescents diagnosed with ADHD and ODD, as well as their parents.  Families were randomly assigned to receive one of two family-therapy treatments; 1) 18 sessions of Problem-Solving Communication Training (PSCT), or 2) 9 sessions of Behavior Management Training (BMT) followed by 9 sessions of PSCT.  A brief description of these 2 treatment options is provided below.
 

Problem Solving Communication Training - The PSCT treatment included three primary components for changing parent-adolescent conflict.  In the problem solving component of the treatment, parents and teens were trained in a five-step problem-solving approach: 1) problem definition; 2) brainstorming for possible solutions; 3) negotiation around these solutions; 4) decision-making processes surrounding a solution; and 5 and implementation of the solution.  This training was intended to help parents and adolescents develop new skills for resolving disagreements with less conflict.  Adolescents were required to attend all 18 sessions of this treatment.
 
The communication-training component focused on helping parents and teens develop more effective communication skills when discussing family conflicts.  For example, parents and teens were taught to maintain an even tone of voice, to demonstrate an understanding of the others' concerns before voicing one's own concerns, to avoid insults and put-downs, and to provide approval for positive communication.  These skills were intended to reduce the use of aversive communication strategies that can make parents and teens angrier, and thereby intensify the conflict.
 
The final component of PSCT was training in cognitive restructuring.  This involved helping families learn to detect, confront, and modify irrational, extreme, or rigid belief systems held by parents or teens about their own or the others' conduct.  This aspect of the treatment was intended to combat the overly rigid and biased views of one another that may develop in families marked by conflict, and which can make resolving conflicts more difficult.
 

Behavioral Management Training/PSCT - In this treatment, the first nine sessions were attended by parents only and were devoted to teaching parents more effective behavior-management skills.  Session topics included: the use of positive attention to promote desirable behavior; developing a point system for reinforcing the accomplishment of responsibilities; using age-appropriate punishments and loss of privileges for undesirable behavior; and teaching parents how to anticipate problem situations and develop plans in advance for dealing with them.  Following this nine-week instruction in behavior management, the teens joined parents for the final nine sessions and the PSCT approach described above was implemented.

 
RESULTS

The researchers collected a variety of measures from mothers, fathers, and teens to evaluate the impact of each treatment.  This included participant ratings of the quality of parent-teen interactions, the frequency and intensity of conflicts, and the strategies used to resolve conflicts when they occurred.  In addition, families were videotaped while discussing a recent situation that had generated conflict, so that their actual behavior during conflicts could be observed and analyzed. Results are summarized below.
 
How many families completed the treatment?
 
Halfway through treatment, 26% of families in the PSCT group had already dropped out, compared to only 8% of families in the BMT/PSCT condition.  By the end of treatment, these numbers had risen to 38% for the former and 18% for the latter.
 
The researchers suggest this may have occurred because, in the PSCT approach, teens attended all sessions and families were immediately required to deal with difficult issues.  In the BMT/PSCT condition, in contrast, parents initially attended by themselves and worked on developing more effective behavior management skills rather than immediately discussing issues of conflict with their teen.  As a result, parents may have developed a greater comfort level with the therapist and more effective strategies for dealing with their child's oppositional behavior before beginning the direct and difficult interaction with their teenager. The researchers suggest this may be the reason why fewer of these parents chose to end treatment prematurely.
 

How effective was each treatment approach?
 
The answer to this question depends on how one chooses to examine the results.  On virtually all outcome measures collected directly from participants, significant improvements were evident. This was true for mothers, fathers, and teens themselves. Thus, participants reported that fewer parent-teen conflicts were occurring, that the anger experienced during conflicts had declined, and that more effective strategies for resolving conflicts were being used.  This was true for families in both treatment conditions, and no significant differences between the treatments were found.  Furthermore, these apparent gains were still evident -- for the most part -- in follow-up data collected two months after treatment had ended, as parents and teens reported high levels of overall satisfaction with the treatment.   These findings are certainly encouraging.
 
A somewhat less optimistic picture emerges, however, when other aspects of the results are considered.  First, given the significant improvements participants reported in multiple areas of parent-teen interaction, one would expect substantial changes in the way parents and teens behaved during the videotaped interactions. However, this was the case.
 
Immediately following treatment, observer ratings indicated that mothers were engaging in significantly more positive behavior and significantly less negative behavior than before treatment began.  For teens and fathers, however, no differences were observed.
 
At the two-month post-test, the positive effects that had been evident for mothers no longer were apparent.  In addition, there was an indication that fathers in the BMT/PSCT group were now less positive and more negative than they were immediately following treatment.  Overall, therefore, the positive reports that participants provided were not matched by changes in their actual behavior -- at least in the samples of behavior that could be collected during these brief videotaped interactions.  This calls into question the validity of the benefits reported by the participants.
 

What proportion of families changed as a result of treatment?
 
The results discussed above describe the average level of change for all families in each treatment.  A different, and perhaps more instructive, way to understand treatment impact is to determine the degree of change that occurred within each family.  The researchers considered this issue in 2 different ways.
 
First, they looked at the percentage of families in each treatment group who demonstrated reliable change.  By reliable change, the researchers are referring to change that is greater than what could reasonably be attributed to chance.  It is based on the concept that all families would be expected to show some change in functioning during the time period over which treatment occurred -- or at least in how they respond to the questionnaires -- and that the changes reported must be greater than what may have otherwise occurred to represent a true treatment benefit.
 
From this perspective, the results are less encouraging.  For these analyses, the researchers focused on those measures that were the primary targets of each treatment. Across three different measures of parent-teen relationship functioning -- parents' perception of the quality of the relationship with their teen, the number of different topics that elicit conflict, and the intensity of anger experienced during conflicts -- reliable change was evident in fewer than 25% of families, based on maternal and paternal reports.  For example, in regards to parents' overall rating of relationship quality with their teen, only about 20% of fathers and 15% of mothers reported improvement that was substantial enough to be considered a reliable change.
 
As a final way of considering the data, the researchers also determined the percentage of families whose scores on the different measures moved from the abnormal range into the normal range during treatment.  These results provide a somewhat more optimistic picture.  At the beginning of treatment, between 3% and 40% of mothers provided ratings of overall relationship quality with their teen, number of issues that generate conflict, and anger intensity during conflict that fell within a non-deviant range.  After treatment, the researchers obtained non-deviant ratings on these measures from between 34% and 78% of mothers.  For fathers, a comparable increase occurred.
 
(Note:  These figures seem more positive than the reliable change results reported above because parents' ratings could move from the deviant to the normal range and still not be large enough to reflect a reliable change.)
 

SUMMARY AND IMPLICATIONS
 
The authors should be commended for presenting their results in such a careful and rigorous fashion.  The overall summary they provide of their results is that "...the findings raise serious questions about the effectiveness of using parents as the major focus of achieving change in adolescents with ADHD/ODD when there is significant interpersonal conflict with parents."  They base this conclusion on the fact that, although significant benefits were found when averaged across families, the percentage of parents for whom "reliable" change was found -- a more conservative approach for estimating the benefits of treatment -- placed them in a distinct minority.  In addition, the observational data were notable for the relative absence of significant improvements in parent-teen interaction during actual conflict.
 
The family-treatment approaches used in this study are well-developed interventions that were delivered for what seems to be an adequate time period.  In fact, the duration of treatment in this study was twice as long as what has been delivered in earlier studies of family treatment for adolescents with ADHD.
 
Why were more positive results -- as indicated by reliable improvement occurring for a greater number of families -- not obtained?
 
The authors suggest several possibilities.  First, they speculate the entire model of having such treatments delivered in a clinical setting, rather than working directly with families in their homes, may have undermined treatment effectiveness.  Working with families at home where their problems actually occur may enhance the impact of the treatments provided.  In fact, there is evidence to suggest that intensive in-home treatment for families where a teen has serious behavior problems can produce positive results.  One exemplar of this treatment approach is called multi-systemic therapy (Learn more about this treatment method at http://www.mstservices.com.)
 
The researchers also suggest, however, that the family may be less important for reducing conflict in families where there is a teen with ADHD/ODD than is commonly believed.  They point to recent evidence suggesting the influence of genetic factors on such aspects of family functioning as parent-child conflict and family cohesion may actually increase in strength as children move into the adolescent years.  They note that, although such findings do not preclude the possibility of inducing changes in parent-teen conflict through efforts to alter communication patterns and parent-management skills, they do imply that parent management of the teen may not be the major source of such conflicts, as many family therapists assume.  If this is correct, the authors suggest that medication treatment may actually be a more helpful approach to reducing parent-adolescent conflict in teens with ADHD/ODD.
 
(Note: It would have been informative if they had considered treatment results according to whether or not the teen was also receiving medication, but this was not included.)
 
This rich and complex data set lends itself to a variety of interpretations, of course, and the ideas put forth by the authors are not the only reasonable way of interpreting their data.  In my own view, they seem to be too pessimistic in their interpretation of the results.  Even using the most conservative definition of improvement -- i.e. reliable change -- treatment was associated with gains in up to 25% of families on some of the measures.  In addition, because they did not employ a control group, the extent to which families would have changed in the absence of any treatment cannot be determined.  Finally, the families themselves clearly felt as though treatment had been helpful to them - a fact which is important to consider.
 
Therefore, interpreting these findings as evidence that family treatment for teens with ADHD/ODD is unlikely to be helpful is not the only way this data can be interpreted.  One can also interpret these results as suggesting these treatments can be of value, and that families who participate in this treatment will find it helpful.  Even this more optimistic interpretation of the findings, however, does not negate the fact that promoting reliable and significant improvements in parent-teen relations when the teen has ADHD and ODD is a difficult task.
 
As discussed above, this was difficult to do, even when high-quality family treatment was provided and when many of the teens that participated also were receiving medication treatment. Developing effective methods for preventing the development of ODD in children with ADHD, and the conflict patterns of parent-child interactions that generally accompany this, is thus an even more important objective for researchers and clinicians to pursue.