My academic career began in 1994 with a PhD on social phobia with Professor David M. Clark. He introduced me to his model, his experimental approach, his therapeutic work, and he guided me faithfully through to completion. During this time, I imbued myself with all the other research on various anxiety disorders: panic disorder, agoraphobia, post-traumatic stress disorder, specific phobia, OCD, and generalised anxiety disorder. There was masses of this research published! And it all seemed to be saying the same thing; anxious people make catastrophic interpretations of their experiences and they habitually do things (avoid situations, worry, suppress emotions) that prevent them from challenging these beliefs.
Forward wind 24 years and I can tell you where all this has led, personally, for me. I am convinced that the core of what makes anyone seek help for any mental health problem is that they are experiencing chronic unresolved conflict between their life goals that has undermined the control they need over their life. Whether a person worries a lot, avoids situation, detects threats around them, this is only a problem if it gets in the way of what they want in life.
So, forget the names of all these fancy cognitive and behavioural processes, and all those psychodynamic defences. A therapist just needs to focus wholeheartedly on their client, and help the client to unravel a way round the counterproductive things they are doing, and the paradoxes of how to balance their priorities, and just keep them focused there for as long as they tell you they need, to make a difference. None of this requires a diagnosis, and none of it is specific to anxiety disorders. I have tried to express this in an accessible way right here, but it is supported by scientific theory and evidence (Alsawy et al., 2014). So, who would have thought that the most ethical, practical and financially efficient way to help people with mental health problems would also have the best fit with the science!
The review by Peter Norton and Daniel Paulus is situated somewhere along this path from a way of providing cognitive therapy that is tied to the professional’s diagnosis of the patient, to a fully universal approach that puts the control over the process of therapy in the hands of the client within a context that stretches well beyond the therapist’s office.
Norton and Paulus have taken a narrative review approach, which means they draw together a wide range of evidence to make a particular case. This is wholly appropriate because the type of evidence they needed to review was so wide. They cover the various important features of anxiety diagnoses (such as whether they are reliable, stable over time, occur together), as well as the genetic, social and parental origins, the psychological processes and brain systems, as well as how easily people can recover from different anxiety disorders with therapy.
If you were to read the abstract only, the review by Norton and Paulus might seem quite uncontroversial. But nothing could be further from the truth.
In essence, they have scoured the literature to provide an incontrovertible case that most of the factors involved in anxiety disorders are shared across these disorders, and not specific to any diagnosis. In fact, they point to evidence that a sizeable proportion of people may shift from one anxiety disorder diagnosis to another every two to six years.
The evidence points to a common genetic predisposition, a common experience of how they were parented as a child, as well as the same psychological processes such as worry and avoidance. It is clear though that the themes of the concerns in different anxiety disorders are different; fears of being rejected and ridiculed in social phobia versus fears of imminent collapse or breakdown in panic disorder, for example. This may explain why the review identified some evidence for differences in the brain systems that are more active in people with different disorders. Norton and Paulus then round up the article with their own transdiagnostic model of anxiety; grounded within the cognitive behavioural domain.
The authors state (p133):
Perhaps most importantly, the preponderance of the epidemiological, diagnostic, genetic, neurological, developmental, cognitive, and behavioral evidence reviewed here suggest far higher degrees of similarity than difference among the anxiety disorders, thereby calling into question the validity, utility, and purpose of retaining a diagnostic substructure within the anxiety disorders. Thus, a transdiagnostic model of anxiety appears not only to be best supported by most (though not all) data at multiple levels of analysis, but also better aligned with the pursuits of our field at the current stage of development.
Norton and Paulus also point out that even though some diagnostic difference are found, these differences were not relevant to their purposes, presumably to understand the nature of anxiety well enough to provide the best intervention to aid recovery. They use a neat everyday example to make their case. We can all see the differences between an Adirondack deck chair and a black leather lounge chair. But these differences are not relevant to the purpose of finding a seating area for guests at a party. If one has a different purpose; to sunbathe on a hot day, then their differences are more relevant. I think this question of ‘purpose’ is absolutely critical, and this is at the heart of what I see as the strengths and limitations of the review.
Strengths and limitations
The scope of the review is its greatest strength. It has ambitiously covered all the areas relevant to anxiety, from whatever your professional interest. I was very familiar with their arguments, but to see some of the most compelling contemporary evidence all in one place, is important. Everything written in the review is clear, evidence-based, and appropriately balanced, and they wrote just enough to suit their purposes. But was the purpose of the review, the most suitable purpose for the field of mental health? I think not, and this is where I turn to what the review left out.
First, to be fair, I think that the reader should get to see the case that is being made by other academics and mental health professionals for a diagnostic approach. What arguments are being made, and what are their evidence? It is possible, echoing the chair analogy, that there are purposes for which a diagnosis is an advantage compared to a transdiagnostic approach. Let’s hear these cases and respond to them.
Second, whilst I can see the strength in focusing on the robust evidence base within anxiety disorders, this constraint does little to advance a truly transdiagnostic approach. Anxiety disorders are one of nearly two dozen different groupings of mental health problems. It seems an awful waste given that my colleagues and I already established the evidence for a transdiagnostic approach across all adult mental health problems 14 years ago (Harvey et al., 2004), albeit within the evidence for cognitive and behavioural processes. Anxiety symptoms are experienced by people with most mental health problems. Even diagnosable anxiety disorders are present in the majority of people seeking treatment with other diagnoses such as bipolar disorder and psychosis.
Third, the review makes an excellent scientific case for the transdiagnostic approach, but what about the practicalities, the economics and the ethics? In my experience, these arguments are at least as strong as the scientific ones. It takes huge amounts of resources to train professionals in recognizing specific diagnoses and learning all the different treatment models. It takes the client’s time to reach an accurate diagnosis, which whilst useful to some, can be stigmatizing to others. And who should really be determining what to talk about in therapy? Surely it is the most ethical choice to leave that up to the client, and not limit it to what is seen as relevant by their diagnosis?
Fourth, the reader is left with a transdiagnostic model that is hard to disagree with from the case made in the review. But do all the differences between the boxes in the model serve a necessary purpose, or is there another review required to bring out what is shared between them? The work that our group has been carrying out suggests that all these boxes can be narrowed down to one core process (e.g. Mansell & McEvoy, 2017). We have used the blind men and the elephant metaphor. Each of these processes are different parts of a whole system (the whole elephant) and psychologists need to understand how these elements work together. We can take inspiration from Charles Darwin, who used one simple theory to explain change across all the categories of living things, despite their obvious differences from one another (Mansell et al., 2014).
Implications for practice
Most of the research supporting this article has been in the public domain for many years. However, bringing it together like this provides maybe the strongest empirical case so far for a transdiagnostic approach to anxiety. What can those in mental health services do to make a difference based on this case? The implications are particularly broad and deep, compared to most research you might read about on the Mental Elf!
- As a clinician, spend more time helping people with the psychological problems and processes that they want to talk about, than what you think you need to establish an anxiety disorder diagnosis.
- As a researcher, teach, train, and supervise transdiagnostic interventions for anxiety to complement and compare with existing approaches. Evaluate all of these and clients’ experiences of them and disseminate these findings.
- As a service-lead or commissioner, treat the disorder-specific NICE guidelines for psychological interventions with a healthy openness; they are not wrong; they are simply science in progress and not a parsimonious summary how to help people with anxiety problems.
Conflicts of interest
My only conflicting interest is my overlap in research interest with the authors and my publication of earlier reviews on a related topic.
Norton PJ, Paulus DJ. (2017) Transdiagnostic models of anxiety disorder: Theoretical and empirical underpinnings. Clinical Psychology Review, Volume 56, 2017, Pages 122-137. [PubMed abstract]
Alsawy, S., Mansell, W., Carey, T. A., McEvoy, P., & Tai, S. J. (2014). Science and practice of transdiagnostic CBT: a Perceptual Control Theory (PCT) approach (PDF). International Journal of Cognitive Therapy, 7(4), 334-359.
Harvey, A. G., Watkins, E., & Mansell, W. (2004). Cognitive behavioural processes across psychological disorders: A transdiagnostic approach to research and treatment (PDF). Oxford University Press, USA.
Mansell, W., & McEvoy, P. M. (2017). A test of the core process account of psychopathology in a heterogenous clinical sample of anxiety and depression: A case of the blind men and the elephant? (DOC). Journal of anxiety disorders, 46, 4-10.
Mansell, W., Carey, T. A., & Tai, S. J. (2015). Classification of Psychopathology and Unifying Theory the Ingredients of a Darwinian Paradigm Shift in Research Methodology. Psycopathology Review, 2(1), pr-036114.