Process‐based and principle‐guided approaches in youth psychotherapy

We appreciate the rich, thought‐provoking paper by Hayes and Hofmann¹, including their inspiring account of the work of so many intervention scientists on whose shoulders we all stand. The directions they propose warrant close attention by all of us who seek to strengthen psychotherapies. Here, we focus specifically on how their ideas may apply to youth psychotherapy and idiographic treatment of youth mental health challenges.

Youth and adult psychotherapy have obvious similarities, but differ in ways relevant to Hayes and Hoffman’s analysis: a) caregivers’ involvement in accessing and participating in their children's treatment highlights the salience of caregiver support and “styles of family functioning”, which Hayes and Hofmann identify as mediators of outcome; b) youths, unlike adults, often begin treatment at the behest of their caregivers and teachers, not for intrinsic reasons, and this can make motivational processes especially critical to success in youth therapy; c) youth developmental stage may impact the accessibility and ef­ficacy of some therapeutic processes (e.g., recursive reasoning about one's own cog­nitions; regulation of attention and emotion through mindfulness and sense of self, prominent in some “third‐wave” therapies).

These caveats notwithstanding, much of the authors’ analysis is directly relevant to youth psychotherapy. For example, they stress that, although psychotherapy protocols have often outperformed comparison conditions, advances in efficacy to date have “been inhibited”. This perfectly characterizes the youth psychotherapy literature. In a recent meta‐analysis², we synthesized findings of 453 randomized controlled trials of youth psychotherapies, spanning five decades. Across time, mean effect sizes have not changed significantly for treatment of anxiety and attention‐deficit/hyperactivity disorder (ADHD), and have declined significantly for depression and conduct problems.

Those worrisome findings were complemented by an analysis of the potential for improvement of current psychotherapies³. Using a meta‐analytic copula approach with 502 randomized trials, we predicted youth psychotherapy effect size as a function of therapy quality. Our results indicated that a currently available therapy of “perfect quality” would have an estimated effect size of Hedges’ g=0.83, conferring (via common language effect size) a 63% chance – only 13% better than a coin‐flip – that the average treated youth would improve more than the average control group youth. This suggests, consistent with Hayes and Hofmann, that truly major improvements in therapy benefit may require fundamental changes in our interventions.

But, aren’t new and different therapies being designed every year? Yes, but the challenge has been to create new therapies that are not skeuomorphic – new in some respects but retaining unnecessary and potentially counterproductive features of their predecessors⁴. Optimizing advances may require both building on strong foundations and breaking the mold. Hayes and Hoffman wisely note the value of leveraging the strengths of existing therapies when innovating, making intervention development evolution, not revolution. We agree. The challenge may lie in striking the delicate balance between incorporating decades of evidence on what works, and shedding structures that are based in tradition or habit, rather than evidence.

Achieving the right balance could involve, as the authors suggest, focusing on change processes and making treatment more idiographic, less standardized. They suggest “moving away from treating psychiatry labels toward treating the individual patient by understanding the process‐based complexity of his/her problems and applying tailored intervention strategies”. Our efforts, and those of our colleagues, to apply such an approach in youth psychotherapy have led to the creation of treatments that are modular, transdiagnostic, and personalized using measurement‐based care. In one version, called MATCH5, 6, 33 components (i.e., “modules”) of evidence‐based treatments for anxiety, depression, trauma, and conduct problems – all derived from decades of research by our predecessors – are organized into a menu of treatment options. Clinicians use this menu to design treatment idiographically, guided by decision tools and an individual dashboard showing each youth’s treatment response, updated weekly. Although decades of research inform its content, MATCH departs from traditions such as treating just one psychiatric disorder and using a standardized sequence of sessions – potential skeuomorphs but, at a minimum, not features that research has shown to be essential for beneficial outcomes.

In a second step of idiographic design, we have organized youth psychotherapy around empirically supported principles of change, honoring ideas previously proposed by many leaders in the field⁷. The resulting FIRST protocol8, 9 synthesizes treatment procedures within five principles: calming and self‐regulation, cognitive change, problem‐solving, positive opposite behaviors (e.g., exposure, behavior­al activation), and motivation for change. This principle‐guided approach rests on the rationale that learning specific procedures is useful, but perhaps most useful to therapists who understand why they are using certain techniques – i.e., which change processes need to be set in motion to produce real benefit. In FIRST, as in MATCH, treatment is fully idiographic, with individualized intervention guided by clinician decision tools and repeated mea­surement of each youth’s functioning and treatment response.

Early evidence on these idiographic ap­proaches has been both encouraging and revealing, highlighting what youth psychotherapy research suggests may be three key challenges for process‐based psychotherapy. One challenge is clinical decision‐making. As treatments become less standardized and more idiographic, clinicians will be required to decide, for each youth, which processes to target, in which order and in which combinations, and with which specific procedures, given multiple options supported by evidence. A critical long‐term task for intervention science will be developing strategies for guiding such decision‐making, and determining the optimal blend of data‐driven and clinician‐guided judgment.

A closely‐related challenge will involve enriching and deepening clinical assessment to capture the underlying processes that need attention in treatment – processes that may be key to therapeutic success. Our field has a long history of assessment focused on diagnosis and symptoms, and a respectable track record within some of the process dimensions identified by Hayes and Hofmann – for example, cognitive reappraisal, rumination, worry, and catastrophizing. However, the newer, deeper, contextually‐focused processes identified by the authors – such as cognitive diffusion, flexibility, non‐reactivity, and “healthy psychological distance from thought” – may well require new measures, and possibly entirely new assessment strategies.

A third challenge will be discerning the implications of process‐based psychotherapy for what many consider the holy grail of intervention science: identifying mechanisms of change. There is a long history in our field, well‐documented by Hayes and Hofmann, of efforts to elucidate mediators of therapeutic change. Documenting mediators is a statistical step toward identifying mechanisms that account for treatment benefit – the switches that, when flipped, make therapy successful.

An implicit assumption historically has been that we will eventually discover the mechanisms of change (or perhaps a small number of them) for treatment of each psychiatric disorder. A process‐based analysis turns this thinking upside down in at least two ways: a) treatment focuses not on disorders but on underlying processes, and b) treatment is tailored to each individual, targeting complex underlying processes that matter for that individual. Under these conditions, do we continue the search for mechanisms of change and, if so, are we searching for “flip switches” as diverse and distinctive as the individuals our interventions are designed to support?

Taken together, there is much that intervention scientists – including those of us immersed in youth psychotherapy – can learn from the perspective offered by Hayes and Hofmann. Clearly, exciting challenges lie ahead in process‐based psychotherapy.