By: Lydia Furman, MD Assistant Editor
The work of Silverstein et al, (doi:10.1542/peds.2014-3221) “Collaborative Care for children with ADHD symptoms: a randomized comparative effectiveness trial,” is highly intriguing for several reasons. The authors studied a low-income inner-city population of children (and parents) that presented with ADHD symptoms, and randomized them to Enhanced or Basic Collaborative Care. Enhanced Care included a trained lay facilitator who met with the family up to 5 times to address barriers to treatment and parental mental health concerns, and to provide ways to reduce coercive parenting in response to difficult behaviors.
The title raises a very interesting question – does diagnosis of ADHD help clinicians manage symptoms of inattention and hyperactivity? The authors’ approach suggests that it may not, since they enrolled all eligible children who were being evaluated for “ADHD symptoms”, and included both those who were ultimately diagnosed with ADHD, as well as those who were not, in the clinical trial. This empiric research strategy encourages us to think more broadly about how to help children who present with symptoms of inattention, impulsivity and hyperactivity. The diagnosis of ADHD as described “was defined as meeting DSM-IV symptom criteria on both the parent and teacher Vanderbilt scales, in the absence of a plausible alternative explanation for the child’s symptoms- as determined from narrative clinical information.” Given this straightforward and pragmatic evaluation using an instrument with relatively low positive predictive value (PPV of 0.19 and 0.32, respectively; Bard et al J Dev Behav Pediatr 2013; Wolraich et al J Dev Behav Pediatr 2013), it is worth considering the possibility that the 63 “ADHD consistent presentation” subjects (40% of all subjects), as compared to the “ADHD inconsistent presentation” subjects, differed quantitatively not qualitatively from each other. In other words, did children in the former group simply have more symptoms than those in the latter group, rather than a unique disease state?
The “ADHD consistent presentation” subjects, whose outcomes were analyzed secondarily since the study was powered to look at outcomes of all subjects, showed greater symptom improvement in the Enhanced Collaborative Care group (please read the paper to learn about the summary results for all children!). Given the study design, the authors could not determine which component of the Enhanced Care was the most helpful. However, since there was not a statistically significant difference in specialty behavior services or in medication treatment between groups (52% in basic care vs. 72% in Enhanced care, p=0.10), and there was a difference in receipt of Triple P (Positive Parenting Program http://www.triplep.net/glo-en/home/) with 0% in basic care vs. 47% in the Enhanced Care Group receiving the intervention, the authors speculate that this could explain the impact of the Enhanced Care. In deference to the authors, I note that they believe the clinically meaningful difference in medication use between groups likely was important also, but we know from multiple publications, including the very well monitored MTA study in which actual medication adherence was 53.5% (Pappadopulos et al Medication adherence in the MTA JAACAP 2009), that compliance with medication treatment may be surprisingly low.
In an accompanying Perspective, Dr. Mark Wolraich,(doi:10.1542/peds.2015-0070) who is the lead author of the AAP ADHD Guidelines and of the Vanderbilt forms, notes that “while progress is being made,” no etiology has been identified for ADHD in over a decade of research, and “therapy is likely to remain symptom based,” which is indeed the approach that Silverstein et al take. Dr. Wolraich also notes that long term outcomes are not yet acceptable, and that even the MTA study found no difference in intensively monitored groups 2 years after treatment ended. In fact at 24 and 36 months post formal intervention, children in the MTA who were taking medication, as compared to those who were not, showed significantly greater symptom deterioration from 24 to 36 months, as well as higher delinquency at both these time points (Jensen et al 3-Year Follow-up of NIMH MTA JAACAP August 2007), information that may be new to many clinicians. Perhaps an entirely new paradigm is needed.